Financial Financial Crises and the Presence of Institutions Foreign Banks Center by Adrian E. Tschoegl

03-35 The Wharton Financial Institutions Center

The Wharton Financial Institutions Center provides a multi-disciplinary research approach to the problems and opportunities facing the financial services industry in its search for competitive excellence. The Center's research focuses on the issues related to managing risk at the firm level as well as ways to improve productivity and performance.

The Center fosters the development of a community of faculty, visiting scholars and Ph.D. candidates whose research interests complement and support the mission of the Center. The Center works closely with industry executives and practitioners to ensure that its research is informed by the operating realities and competitive demands facing industry participants as they pursue competitive excellence.

Copies of the working papers summarized here are available from the Center. If you would like to learn more about the Center or become a member of our research community, please let us know of your interest.

Franklin Allen Richard J. Herring Co-Director Co-Director

The Working Paper Series is made possible by a generous grant from the Alfred P. Sloan Foundation

Financial Crises & the Presence of Foreign Banks

Adrian E. Tschoegl Adjunct Assoc. Prof. of Management The Wharton School of the University of Pennsylvania SHDH 2000 Philadelphia, PA 19104 Email: [email protected]

5 December 2003

Prepared for and presented at the World Bank conference: Systemic Financial Distress: Containment and Resolution 7-8 October 2003

I would like to thank G. Caprio, P. Honohan, and L. Laeven for helpful comments on an earlier draft and G. Noguera for research and editorial assistance. At the conference L. Goldberg and M.S. Martinez Peria provided incisive discussant’s comments. Participants at the conference including C. Calomiris, E. Rosengren and also provided helpful comments. 2

Financial Crises & the Presence of Foreign Banks

Abstract: The paper distinguishes between the classic or traditional foreign banks with their emphasis on corporate and wholesale banking, and the innovators responding to transition, deregulation or crisis in Emerging Markets. The innovators come in three varieties— bettors, prospectors and restructurers—with their role depending on their type. The paper argues, on the basis of 12 short national or regional case studies, that foreign banks have had little effect in eviting crisis, in great part because generally they were not present in any scale before the crisis. In the recovery phase foreign banks can act as rehabilitators of weak or failed banks, and can help ward off future crises by taking banks out of government or family ownership. The paper’s most controversial argument is that to the degree that reform succeeds, the conditions that attracted the foreign banks disappear and the domestic banks are able to grow more rapidly, in subsequent decades many foreigner owners are likely to sell their subsidiaries to local banks and investors. Thus in the succeeding decades, the ratio of assets in foreign-owned banks to total banking system assets should decline, even in the absence of government policies that aim for that result.

1.0 Introduction

Foreign banks have entered many transition and emerging economies in recent years, sometimes before economic and banking crises have developed, and often after. Today, in a number of countries foreign banks own as much as 90 per cent or more of the banking systems’ assets. The question then naturally arises as to what the effect of the foreign presence is on crises, especially on preventing or ameliorating them.

The paper begins by discussing the motives, modes and regulation of foreign banks.

In analyzing foreign entry one must distinguish between classic foreign banks and the innovators that we will refer to as bettors, prospectors and restructurers. The paper argues that the innovators enter in response to the opportunities that crisis creates, but that the

banks’ entry and the reforms that accompany it erode the very conditions that drew the

banks. In the long-run then, one can expect their relative and possibly even absolute

importance in host-country banking systems to shrink. Next, the paper distinguishes entry

via branches and via subsidiaries as the mode is tied to the motive. In the third section, the

paper reports that the evidence suggests that foreign banks have had relatively little effect in

crises, primarily because they did not have a significant presence before the crisis began. 3

Lastly, the paper argues that foreign banks can help in the aftermath of a crisis by acting as

micro-economic rehabilitators of failed banks and as instruments of macro-economic

reform. However, the foreign banks come with their own limitations. The paper also

includes 12 appendices that discuss the role of foreign banks in crises in a selection of

countries.

2.0 FDI in Banking in Emerging & Transition Economies

To provide some background to the discussion of the role of foreign banks in

financial crises, this section deals with motives, modes and regulation. Motives deals with

the strategies of the foreign banks that bring them to the host countries. Modes deals with

the legal form of the banks’ presence. Lastly, regulation is critical to understanding the

scope of activities in which foreign banks may engage.

2.1) Motives

One can broadly classify foreign banks into two categories based on their strategies

in the host countries: the traditionals and the innovators. The traditionals consist of those many foreign banks, especially those operating via a branch or a small, wholly-owned subsidiary in a national financial center, that are essentially engaging in normal international banking. These banks process trade payments, finance trade, trade foreign exchange, and lend to corporations, both to their home country clients operating in the host country and to host country firms that require an alternative to the services that local banks can offer. The traditionals do not seek to engage in retail banking, with the limited exception of ethnic banks, that is, banks that have come to serve the needs of recent or perhaps even long- established emigrants from the home country resident in the host country. 4

Deregulation, transition and crisis have created opportunities for the second group,

the innovators. The innovators are innovative in two, non-exclusive ways: frequently their

responses to the new opportunities represent behaviors that are new to the banks

themselves, and their responses are to bring governance, methods, and products that are

new in the markets that they enter, or at least scarce. They also bring capital, but this is in many ways their least important function. The innovators come in three varieties, the bettors, the prospectors, and the restructurers.

The bettors include everything from various quangos (quasi-autonomous, non- governmental organizations) and development banks to banks, investment or private equity firms and individuals betting on the post-deregulation environment or on the success of transition. Some examples would include the International Finance Corporation, the

Deutsche Entwicklungsgesellschaft, and the European Bank for Reconstruction and

Development. The bettors’ involvement is heavily financial. Through their activities they acquire a portfolio of investments. They hope to create the bulk of the value to themselves through having made good bets on the capabilities of the institutions and the management teams in which they invest. In time one can expect them to exit due either to the success or to the failure of their investments. In the meantime they may perform a screening and signaling function. Their involvement in banks as major investors (10-40 per cent share positions) gives the banks in which they invest some legitimacy in an environment where there may be numerous new and reorganizing local banks.

What distinguishes the bettors from the prospectors and the restructurers is that the bettors do not actually manage the banks in which they invest. The bettors’ governance role is frequently a negative one–they impede malfeasance. Their presence therefore is a signal of probity but because they do not manage their investment, less so one of competence. The 5 bettors invest together with a strategic investor that has the largest and often a majority stake. A strategic investor is a firm, usually operating in the same or a similar field, which takes a controlling equity stake. The strategic investors acquire their targets and intend to manage them. These strategic investors do expect to earn a return on risk capital, but more importantly they also expect to earn a return on the management skills, systems and experience that they bring to bear.

The structure of strategic investor supported by bettors is similar in practice if not in legal form to the Kommandite of the pre-Great Depression world. The strategic investor is analogous to the general partner in a Kommandite and the bettors are analogous to the limited partners. Strategic investors may be domestic or foreign, and the foreigners in turn may be prospectors or restructurers.

The prospectors are foreign banks that perceive the existence of a possible opportunity in the unsettled situation following opening, crisis or deregulation. The prospectors typically have no particular experience with the host country or clients there.

The prospectors simply are willing, often for idiosyncratic reasons, to engage in exploratory forays. What distinguishes prospectors from restructurers (see below) is that the prospectors’ engagement is exploratory rather than part of a broader strategy.

The prospectors establish small subsidiaries or joint ventures, or take large equity positions in local banks that are small relative to the foreign bank. If the foray shows promise they are willing to expand further. If it does not, they cut their losses, generally by selling out to local or other foreign banks. The metaphor here is the gold mine model of

Moffett et al. (1989). The prospector sinks an exploratory shaft in a promising location. If he does not strike gold-bearing ore, the prospector moves on. If he does strike gold the mine then follows the ore vein, wherever it leads. One illustrative example is Allied Irish 6

Banks (AIB) and its investments in Poland and Singapore (Table 1a). A more complex example involving several prospectors is Prager Handelsbank (Table 1b).

Many prospectors establish de novo ventures because they have very focused strategies. Examples would include Porsche Bank and Opel Bank in Hungary, both of which specialize in consumer finance, especially the purchase of their owners’ cars. The evidence in Majnoni et al. (2003) for Hungary suggests that greenfield operations achieve higher profitability than restructured ones, and that this is due to the introduction of new financial products.

The last category is that of the restructurers. These acquire a large bank in a privatization or a rescue and proceed to attempt to fix it. The investment is not a trial foray but rather a large-scale commitment. Frequently, it is part of a program of similar investments. Banks such as Unicredito (Italy), Erste Bank (Austria), KBC (Belgium), Société

Générale (France), and others have acquired banks throughout the transition economies. As one can see from Table 2, these restructurers have acquired banks in four or more countries.

From the names of the local banks alone one can infer that these are recent acquisitions.

(The names differ from country to country and the parent has not yet rebranded the subsidiaries with a common identity.) Also, in several countries these four are competing with each other; in other countries, though only one to three of these banks is present, other foreign banks are also present. The presence of several foreign banks from different countries brings a rivalry to the market for bank services.

Although the identities of the restructurers may change, one can find similar patterns in other regions of the world. Nordic banks have been active in the Baltic. Greek banks such as National Bank of Greece, Commercial Bank of Greece and Alpha Bank have been equally acquisitive in the Balkans-Black Sea region. Somewhat similarly, banks such as 7

Santander and BBVA (Spain) and HSBC (UK) have acquired banks throughout Latin

America (Guillén and Tschoegl 2000).

Although some prospectors and restructurers may hope to sell out within the medium term (say 3-7 years), most view their engagement as open-ended. At entry they have no exit plan but instead anticipate retaining ownership into the indefinite future, subject of course to profitability. However, in time one can expect that most will sell their operations to local banks as the evolution of the host markets erodes their comparative advantage vis-à-vis their local competitors.

The basis for the argument for the long-run return of the financial systems to predominantly domestic control is the metaphor of ecological succession (Koford and

Tschoegl 2002). Ecological succession is a dynamic model that posits that the first plants to arrive in disturbed or clear soil stabilize the soil and change light conditions, creating opportunities for successor species. This in turn provides opportunities for yet other species. The mix of plant and animal species continues to evolve until the system reaches the final steady state of a climax forest. Of course, when fire, flood or landslide clears a patch in the forest, the process starts again in that patch. During the process one must be careful not to assume that the current situation is in equilibrium. One should not mistake a current phase, especially not an early one, for the ultimate steady state.

The foreign banks are the fast growing plants that are the first responders to the new opportunity given by the disturbed, hospitable soil that crises or transition have created.

This does not mean that all foreign banks are innovators; some international banks will remain with traditional international banking. It just means that among the foreign banks there are those that will respond quickly and in ways that take them beyond traditional international banking. They can respond quickly because they are well-capitalized, efficient, 8

and have surplus managerial resources (Kindleberger 1969 and Tschoegl 2004). It is also

important to remember that the ecological succession metaphor concerns itself with the

evolution of a system after a disturbance. It is not designed to deal with steady states. It is

not a model for the role of foreign banks in mature, stable markets, though as the case of

Norway shows (Appendix 9) shows, even in mature markets it has applicability in crises. For

a discussion of the role of foreign banks in a large, competitive market such as the U.S. see

Tschoegl (2001).

We can think of the share of banking system assets in foreign banks relative to total

banking system assets as the proportion of the system’s biomass accounted for by the fast

responding plants. One feature of the ecological succession process is that there is both a

numerator and a denominator effect. The behavior of the foreign banks themselves has a

large effect on the numerator. Over the long term the response of the domestically owned

banks will have a large effect on the denominator.

As the foreign banks enter, especially when they acquire host country banks in

privatizations, rescues and repairs, the numerator grows more rapidly than the denominator.

What draws the foreign banks is the opportunities the situations present, now that they are

able or even encouraged to enter. Weill (2003) found that for the Czech Republic and

Poland that foreign-owned banks were more efficient than domestic-owned banks but that

this was not due to scale differences or the structure of activities. Demirgüç-Kunt and

Huizinga (1999) and Claessens et al. (2001) found that foreign banks tend to have higher

margins and profits than domestic banks in developing countries, but that the opposite holds

in industrial countries. In a similar vein, Dopico and Wilcox (2002) found that foreign banks are more pervasive in countries where banking is more profitable and where the banking 9 sector is smaller relative to GDP. Conversely, foreign banks have a smaller presence in mature, competitive markets.

Although what draws the foreign banks are markets that are growing, underbanked or uncompetitive, by their entry the foreign banks change the environment. The literature on the effect of foreign banks on host country banking systems, especially margins, suggests that the effect is precisely to undermine the conditions associated with the presence of foreign banks. For instance, Lensink and Hermes (2003) find that foreign entry is associated with shrinking margins in developing countries, but not necessarily in developed countries.

For the Philippines Unite and Sullivan (2003) found that foreign competition compelled domestic banks to be more efficient, to focus operations, and to become less dependent on relationship-based banking practices.

This means that over time, measured in decades, the foreign banks will find the conditions that drew them have eroded, and some will withdraw, selling their subsidiaries to locally-owned banks. When one examines histories of particular banks or of banking systems in particular countries, one can rapidly identify a number of mechanisms that operate to reduce the foreign presence. Some foreign banks sell out because they find they are not competitive. Others sell because their parent is in difficulty at home and sells the foreign operation in order to raise funds. Yet others sell out to local investors and banks when host country markets are depressed and the foreign owners see little benefit from staying. Lastly, some simply sell out to local banks growing by acquisition.

For instance, in 1999 Sumitomo Bank sold the subsidiary it had established in

California in 1952 to Zions Bancorp, which was expanding into California. Of course, sometimes the exiting bank will sell to another foreign bank. Thus UFJ Holdings (the company created in Japan to merge Sanwa Bank, Tokai Bank and Toyo Trust) announced 10 the sale of United California Bank to BancWest, a subsidiary of Banque Nationale de Paris

Paribas, which had entered in 1970. Sanwa had established its subsidiary in 1972 and Tokai its subsidiary in 1974. In both the Sumitomo and Sanwa-Tokai cases the parents sold not because of difficulties with their California operations but rather because of difficulties at home. Of the eight subsidiaries that Japanese banks established in California between 1952 and 1978, three remain, the rest having disappeared through merger into the survivors or sale; assets in California subsidiaries of Japanese banks peaked in the early 1990s and has fallen since though the largest, Bank of Tokyo-Mitsubishi’s Union Bank of California continues to grow (Tschoegl 2003b).

Similarly, in 2003 Banco Bilbao Vizcaya Argentaria sold its Brazilian operations to

Bradesco. BBVA had bought Banco Excel-Economico in 1998 for 1 real. BBVA sold because it realized that it would be too expensive to achieve a profitable scale. As part of the sale price BBVA took a small equity stake in Bradesco, thus converting its strategy for Brazil from that of a restructurer to that of a bettor. In 2003, Lloyds TSB sold its Brazilian operations to HSBC (Hongkong and Shanghai Bank Corporation). HSBC had entered in

1997 and the purchase of Lloyds is its third in Brazil. Lloyds’ association with Brazil dated back to ancestors of the Bank of London and South America (BOLSA), which in 1863 and

1864 were the first banks in Brazil. However, Lloyds management has undertaken to focus on its operations in the UK. It has also just (October 2003) agreed to sell National Bank of

New Zealand, with which it has been associated since the 1870s, to ANZ Bank.

To this point the discussion has essentially focused on the numerator of the ratio of assets in foreign-owned banks to total assets in the banking system. There is, however, a denominator effect as well. The argument here comes from Huang and Di (2003). They point out that the conditions that constrain foreign firms may also constrain host country 11 firms. If, as we have argued above, the conditions constrain host country firms more than they do foreign firms, the foreign firms may increase their ownership of assets in the host country. However the liberalization and regulatory reform that accompanied the entry of the foreign firms benefits the domestic firms as well. It will take time for them overcome their administrative heritage from the previous conditions but once they do, one can expect them to grow rapidly. The result will be an increase in the numerator, and a decline in the ratio, even without denominator effects.

Furthermore, in the long-run foreign banks in general have no comparative advantage in retail banking vis-à-vis host country banks, especially in reaching rural areas and the unbanked. Even when the foreign banks are better than local banks at both wholesale and corporate banking on the one hand and retail banking on the other, the local banks will have a comparative advantage in retail banking. That is, the local banks will be less worse at that than at foreign and corporate banking. Thus even when one observes cases where foreign banks initially dominate a country’s banking sector, one should expect to see host country banks come to dominate retail banking. The foreign banks specialize in foreign activities pari passu with the emergence or growth of local competitors, even if they do not immediately exit retail banking. Still, it is not unusual to see a foreign bank sell its retail activities to a local bank while it retains a branch in the national financial capital to enable it to continue to offer corporate and wholesale banking services.

Saying that over time one should expect to see foreign banks exiting retail banking abroad does not mean that all foreign banks will do so. Currently, HSBC is striving to become, in the words of its slogan, “the world’s local bank” (Tschoegl 2003b). Some particularly capable foreign banks may retain a presence for a long time. However, one must 12

keep in mind too that over time management changes, and HSBC’s push in North America

is only about 20 years old, while that in Latin America is about five.

One can argue that “Globalization” makes the metaphor of ecological succession

suspect. However, Goldsmith (1969, 360-367), referring to the 19th and early 20th Centuries,

the first era of globalization, noted that outside of Europe and North America the

development of financial systems was widely based on foreign banks but that over time the

foreign banks had lost their predominance as local banks arose and waxed. Goldsmith does

not go into detail; his focus was on macroeconomics. However, Trainer (1996 and 2002) has

a good description of the development of banking and the banking sector in Brazil,

especially after the reforms of 1905, which shows a pattern consistent with the ecological

succession metaphor. Koford and Tschoegl (2002) apply the metaphor to the history of

foreign banks in Bulgaria since the late 19th Century. Tschoegl’s earlier papers on the foreign bank presence in Japan (Tschoegl 1988) and Saudi Arabia (Tschoegl 2002c) and Phylatkis’

(1988) paper on the eclipsing of British banks in Cyprus before independence, also offer evidence for the applicability of the metaphor.

2.2) Modes

The three most common forms of legal presence are branches, affiliates, and subsidiaries, though their availability to foreign banks varies by country. That is, some countries permit foreign banks to operate via a branch but not a subsidiary, or vice-versa.

Legal form is not an arbitrary formality but is intimately tied to both the banks’ strategies and to the regulatory environment. The distinction between the three is not a matter of arcane legal minutiae but rather of substantive importance, especially in crises. 13

A branch is an integral part of the parent. A branch may make loans or take deposits; generally it may provide a full range of banking services. Banks prefer to use foreign branches for wholesale and corporate banking activities in host countries, including foreign exchange and money market trading. The reason is that a branch generally lends, borrows and trades on the basis of its parent’s full capital base. Thus the branch can lend more to any one borrower than could a similar sized subsidiary (see below) and, in borrowing and trading, the branch shares the parent’s credit rating. Because it is part of the parent bank, a branch requires careful supervision as unauthorized trading at a branch could bankrupt the parent.

An affiliate or associate is an independent legal entity (i.e., locally incorporated) in which the foreign bank has less than majority ownership. Generally, foreign banks prefer not to put their name on affiliates as that would suggest full responsibility for an entity over which they do not exercise full control. Furthermore, if through a combination of share ownership and management contract the foreign bank does exercise control, then it will wish eventually to assume full control to capture more of the return from its management without having to dispute transfer pricing decisions with minority shareholders.

A subsidiary, like an affiliate, is a separate legal entity incorporated in the host country, but one in which the foreign parent has majority ownership. A subsidiary may fail even though the parent is solvent. Conversely, a subsidiary may be solvent even though the parent has failed. The subsidiary lends based on its own capitalization rather than that of the parent, something that can prove to be a major constraint. Certainly, a subsidiary can refer to its parent a loan that exceeds its allowable lending limits but, in that case, why not lend from a branch in the first place? Thus a subsidiary provides a relatively poor platform from which to conduct corporate lending or trading activities (Heinkel and Levi 1992). 14

Generally, a branch costs less to establish than a subsidiary. There are no costs of incorporation, no need to report annually or quarterly to local registrars of companies, no need for a board of directors, and so forth. However, when a bank has sufficient assets at risk where the risk is location-specific there may come a time when it becomes sensible to incorporate locally.

Du (2003) has pointed out that multinational companies may borrow from local banks to enlist the services of those banks as monitors of the local operation. The local bank, as local creditor with often first rights to any assets in bankruptcy, has an incentive to intervene quickly if it senses problems. Similarly, a foreign bank with a number of branches in a country thousands of miles from home, whose profitability depends on a local economy about which the parent knows little, may decide to incorporate locally to enlist the governance services of the local central bank, and local depositors when there is no deposit insurance. These parties are more likely to sense and react to some problems of which headquarters, dependent on reports from its managers, might be unaware. Thus, when the assets at stake are large enough and the local regulatory authorities are capable, the foreign parent may decide that the costs of incorporation are an acceptable cost for improved governance.

As the US deregulated banking, many multi-bank holding companies merged their subsidiaries, converting their offices into direct branches of the parent (Wheelock and

Wilson 2002). The argument above suggests one reason they could do so was because as all were already subject to oversight by the Federal Reserve and Federal Deposit Insurance

Corporation, amalgamating did not imply any loss of a local regulator. By contrast, the little cross-border acquisition activity within the EU has led to little or no amalgamation. That which has occurred has occurred primarily within the Nordic countries. 15

Interestingly, the ten banks that control the 12 largest subsidiaries of foreign banks in the US, which account for about 90 per cent of the assets in subsidiaries of foreign banks, continue to choose the subsidiary form rather than branches of the parent (Tschoegl 2002b).

The parent banks involved do have branches, but these are in financial centers and engage in wholesale and corporate banking. They are not retail banking operations (Tschoegl 2001).

HSBC Canada, the largest foreign bank and the seventh largest bank in Canada, continues to operate via a subsidiary even though it now has the option to adopt the branch form, and even though several foreign banks with only one office in Toronto switched when the option became available in 1999.

It is important to emphasize that this argument for incorporation has nothing to do with taxes. A subsidiary, especially one that the parent does not own outright, has much less flexibility in its tax management than a branch. Lastly, when one observes a parent maintaining, say in the US, both a branch for corporate and wholesale banking and a subsidiary for retail banking, this argues that taxes are not the issue and that the branch’s lack of local limited liability is important for trading.

The choice between operating via a branch or via a subsidiary also may have little to do with the value of the option to abandon implicit in limited liability (see below).

Incorporation as a device for enlisting the governance assistance of host country regulators is a much more promising explanation.

Regulators in the home and host country frequently try to retain the advantage to them of one form or the other. Thus regulators in the home country may attempt to “ring fence” the parent’s assets to insulate them from developments in bank’s branches abroad.

At the same time regulators in the host country may require a branch to allocate a certain amount of capital to the branch to protect local depositors in the event the parent fails. 16

Local regulators may also require a parent bank to issue a “comfort letter” pledging support

to a subsidiary. Still, as we discuss in the next section, regulators monitor and regulate

branches differently from subsidiaries.

The issue of the choice between branch and subsidiary can have implications for the

role of foreign banks in a crisis. In a crisis, and absent deposit insurance, local depositors

frequently engage in a “flight to quality”. That is, they withdraw their funds from domestic banks in which they have lost confidence and redeposit the funds in less risky banks. Often these are government-owned banks, but often these are foreign banks. Depositors will find the branches of a foreign bank, especially one from a country with good regulatory supervision, particularly attractive. Branches cannot fail unless the parent fails, and the probability of the parent failing generally does not depend highly on conditions in the host economy. Also, most parents of foreign branches are large relative to the size of the branch.

Still, although shifting deposits to branches of foreign banks may be a sensible strategy, it is usually of limited use to most citizens because of the paucity of accessible branches. Most foreign banks have only one branch, which they have placed in the host country’s financial center. Obviously the option is of some use to corporations.

Subsidiaries and affiliates may benefit from a flight to quality, but if so, this will depend more on the quality of the management of the bank than on the bank’s legal status.

Both subsidiaries and affiliates are host country legal persons and their performance is heavily bound up with domestic politics and economics as the Argentine case shows

(Appendix 1). A subsidiary may benefit from the perception that the parent will stand behind it to protect the parent’s name and reputation. Obviously, this perception will be much weaker in the case of an affiliate, especially one that has not advertised its links to the foreign parent. 17

In either case, the expectation of parental support is a bet and not a consequence of legal form as is the case with branches. For the parent, the existence of limited liability is an option to abandon with the strike price being the loss of reputation if it walks away.

However the strike price is stochastic in that it depends on the reason behind the parent invoking limited liability. The cases of Crédit Agricole, Scotiabank and MBK Mercobank in

Argentina are illuminating. In all three cases, the foreign shareholders or parents, unwilling to recapitalize failed subsidiaries, turned them over to the Argentine government for rescue.

In the cases of Crédit Agricole and Scotiabank the reasons for failure were the Argentine government’s appalling economic policy decisions. This is a very different matter than if the reason had been corporate mismanagement or malfeasance. Scotiabank did make a voluntary payment of 20 cents on the dollar to holders of its subsidiary’s marketable securities. It may have done so to shore-up its reputation in the global market for its debt.

Regardless of the arguments, the Appendices suggest that the phenomenon of a flight to quality does not provide foreign banks with a long-term advantage. Market share shifts to foreign banks (absent acquisitions of local banks) appear to be temporary and amount to only a per cent or two of the banking systems’ total assets.

2.3) Regulation

The Basle Concordat of 1975 (amended in 1983), established that home country supervisory authorities are responsible for solvency supervision of the parent’s branches abroad. This is logical as it is the home country authorities that have legal access to the parent’s books, and because a branch fails when its parent fails. Host and parent country supervisory authorities are jointly responsible for solvency supervision of subsidiaries, with the host country having primary responsibility. Again this is logical as the subsidiary is a 18

legal person of the host country and so the local authorities have access to its books. Also, a

subsidiary can fail even when the parent is solvent. However, the home country authorities

are responsible for supervision on a consolidated basis as subsidiaries affect the parent’s

solvency and the parent cannot disclaim all responsibility for its subsidiaries. This provision

for joint responsibility followed the collapse of Banco Ambrosiano in 1982.

When Banco Ambrosiano collapsed, the Italian authorities protected Italian

depositors by transferring the bank's business to a new Italian entity. However, they

disclaimed responsibility for the obligations of Ambrosiano’s Luxembourg subsidiary and

the Latin American subsidiaries.

When BCCI failed in 1991, the local regulators in such countries as Canada and

Mauritius dealt with the local operations. The Bank of Canada closed Bank of Credit and

Commerce Canada. Bank of Mauritius issued a license to the Somaia group to establish

Delphis Bank to take over BCCI’s banking business. When Demirbank failed in Turkey in

2000, its subsidiary in Bulgaria continued to function and there was no run on the bank.

Instead, the Bulgarian subsidiary was simply an asset that the Turkish authorities sold.

Again, countries differ in terms of whether they permit foreign banks to operate as

branches, affiliates or subsidiaries. When banks use affiliates, frequently this is in response

to host countries’ law limiting foreign ownership in the banking sector. The laws force the

foreign parent into minority ownership as a condition of entry or continuation of operations.

Generally foreign banks convert affiliates to subsidiaries when they can. Still, in some cases foreign banks continue to maintain affiliates rather than subsidiaries, perhaps in order to limit the parent’s responsibility. Belgolaise, a subsidiary of Belgium’s Fortis Bank that specializes in banking in Africa, generally operates through affiliates rather than subsidiaries. 19

Countries that have wished to open their banking sectors to foreign banks sufficiently to avoid reciprocity concerns while limiting the foreign banks’ scope, have opted to permit subsidiaries rather than branches (Tschoegl 1981). Canada would be a case in point, at least before the NAFTA treaty forced change. Countries that have wished to limit foreign banks to the wholesale and corporate markets and to keep them out of the retail market have permitted branches but not subsidiaries (Tschoegl 1981). Japan would be a case in point, at least before the need to recapitalize some large, failed banks forced change.

In addition to these basic considerations, countries have also had to decide whether or not to permit foreign banks to buy local banks. Even in cases where the country permits the foreign bank to establish a subsidiary it commonly still forbids it to acquire local banks.

This has the effect of limiting the foreign bank to organic growth, something that can be quite slow. HSBC entered Canada in 1979 but only started to grow rapidly in 1986 when it acquired the failed Bank of British Columbia; HSBC Canada followed this with seven more acquisitions, some of which had themselves acquired a total of six banks (Tschoegl 2003b).

Similarly, the foreign banks that own the largest US subsidiaries or affiliates of foreign banks have all grown by acquisition (Tschoegl 2002b). The same appears generally to be the case in those transition and developing countries where there are large foreign banks. The primary exceptions are those cases where the foreign banks came when a country was a colony or before there were locally-owned banks. In these cases, the foreign banks have had the time to grow into a substantial presence (e.g., Jamaica—Appendix 6 and the Pacific

Islands—Appendix 10), though mergers and acquisitions have occurred there, especially among the foreign banks (e.g., Malaysia—Appendix 7).

20

3.0 Effect of foreign bank presence in crises

In general, foreign banks have had little effect in crises, primarily because they did not have a material presence before the crisis began. Of the 12 countries in the appendices

(treating the Pacific Islands as one country), in eight (Bulgaria, Czech Republic, Hungary,

Indonesia, Mexico, Norway, Tanzania and Thailand), the foreign banks’ share of banking system assets ranged from nil to well under 20 per cent. In four countries (Argentina,

Jamaica, Malaysia and the Pacific Islands), it was in excess of 20 per cent.

One could well hypothesize that countries that had the sort of policies that would encourage the entry of foreign banks are not likely to have the sort of policies that lead to crises. Beck et al., (2003) find that crises are less likely in economies with fewer regulatory restrictions on bank competition and national institutions that encourage competition. At the very least, the empirical evidence suggests that developing-country banking systems that have lower entry barriers to foreign-owned banks gain directly and indirectly. They are less vulnerable to crises and the domestic banks are more efficient than in countries with greater barriers. Barth et al., (2003) find that the negative correlation with the likelihood of a banking crisis is not with foreign bank ownership per se, but rather with limitations on foreign bank entry and ownership. Independent of actual foreign bank entry, the authors find that the likelihood of a major banking crisis is positively associated with greater limitations on foreign bank participation.

Even in those cases where foreign banks have had a large presence and there has been a crisis, such as the Pacific islands in the Australian sphere of influence (Appendix 10) or Jamaica (6), the crisis has been limited to the need to bail-out depositors in the government-owned bank. The knock-on effects have been limited. All depositors in the foreign banks have been unaffected and depositors in the government bank have at most 21

only lost temporary access to their deposits if that. There has been no material credit

restriction because the reason for the failure of the government bank was that its lending

went to unproductive purposes. The net effect of the crisis has been simply to transfer

money from taxpayers to the recipients of the political lending of the failed bank.

The great exception has been Argentina. Here part of the problem has been the

asymmetric pesofication that acted as a tax on bank capital and transfer to depositors,

leading some foreign banks to pull out. (For fuller treatments of the Argentine case see

Calomiris et al., (2003) and Honohan (2003).)

The foreign banks tend to have a stabilizing effect on the economy to the degree that

they are present. Subsidiaries of foreign banks tend not to be as affected by crises as the

domestic banks, in part because they have been more conservative in their lending. Dages et

al., (2000) concluded from their study of foreign banks in Argentina and Mexico over the

1994-99 period, that foreign banks exhibited stronger and less volatile loan growth than

domestic banks. However, it was the asset quality of bank portfolios and not ownership per

se that appeared to be the decisive factor. Branches of foreign banks also benefit from

having specialized in lending for trade, especially export trade, and in lending to foreign and

large corporations. Because crises often result from or cause a depreciation of the host country currency, exporting firms benefit from the crisis. Thus, though foreign banks do suffer in from the general downturn that occurs in economies in crisis, they often suffer less than do domestic banks.

In addition, branches certainly and subsidiaries probably benefit from the parents’ support. This has two linked consequences. First, when the foreign banks benefit from a flight to quality this reduces the demand for foreign exchange for flight capital and the 22 consequent pressure on the exchange rate. Second, this gives the foreign banks the funds to continue to lend to host-country firms, which reduces the credit crunch effect of a crisis.

When domestic banks are weak, the subsidiaries of foreign banks may have better access to funding in foreign exchange, especially in times of stress. However, the evidence suggests that strong domestic banks can do as well or even better than the foreign subsidiaries (Reynoso 2002).

Again, the primary constraint on these positive effects is the relatively small presence of foreign banks. A second constraint is that while the country is undergoing a crisis, risk managers at the foreign banks’ headquarters may require their banks to curtail lending. The need to match the currency of the assets to those of the liabilities will then have the foreign banks purchasing government bonds. This probably results in a trivial reduction in the governments’ funding costs but may crowd out private borrowing.

There is one last point about stabilization that is worth mentioning. The reverse of the coin that foreign banks may be less affected than host country banks by problems in the host country economy is that the foreign banks may be more affected by problems in the home country. Both Peek and Rosengren (2000) and Williams (1996), for instance, have documented how problems at home resulted in Japanese banks curtailing their lending abroad. Here the effect was more pronounced with branches than subsidiaries. Capital adequacy issues at the parent affect the ability of branches to lend but do not affect subsidiaries in the host countries.

4.0 Roles of foreign banks in resolving crises

The experience of the 1990s brought an ever-widening recognition of the advantages of inviting foreign banks in (Buch 1997). Foreign banks can operate in two different roles: 23

in the micro-economic role of rehabilitators of troubled individual banks and in a macro- economic role as instruments of reform of the banking system. However, foreign banks also have their limitations.

4.1) Micro-economic rehabilitators

The restructurers acquire failed or laggard institutions that they then rehabilitate.

The transformation process typically begins with the foreign acquirer recapitalizing the acquisition and includes both the closing of unprofitable branches and operations (and the redundancy of staff), and the introduction of systems and processes, especially what is known as “a credit culture”.

As mentioned earlier, frequently governments that have permitted some entry have still blocked foreign banks from acquiring control of domestic banks. It is not unusual for the governments to remove this restriction in crises. Foreigners frequently are the only parties able to recapitalize troubled banks as the other domestic banks are themselves not strong and the government wishes to limit its expenditures where it can. Examples of this abound; see for instance the cases of Mexico (Appendix 9), Norway (Appendix 10),

Tanzania (Appendix 11), and Thailand (Appendix 12).

As Claessens and Lee (2002) have found, foreign banks introduce improved risk management practices and “imported” bank supervision from parent country regulators.

Also, as Lwiza and Nwankwo (2002) point out for Tanzania, and Guillén and Tschoegl

(2000) for Argentina, Chile and Mexico, foreign banks frequently compete by introducing new financial products and services. That is, they introduce to the host country products and services that they have previously developed and deployed elsewhere. The usual 24 microeconomic logic of the gain in consumers’ surplus from new products applies. The introduction of new products also appears to enhance profitability (Majnoni et al., 2003).

As worthwhile as these contributions are, they have little effect on the likelihood of crises. Competition between banks will tend to transfer to customers any efficiency gains.

Somewhat counter-intuitively, even better risk control will probably not affect the likelihood of crises. It may reduce the incidence of failure at individual banks, but examining bank failure in recent crises suggests that the problem has not been the banks’ lack of ability to manage risk but rather a lack of will. Then improved risk control and any reduction in the incidence of crises are both manifestations of the effect of foreign banks on governance.

4.2) Instruments of macro-economic reform

Foreign banks act as instruments of macro-economic reform when governments admit them in order to change the structure of the banking system. There are two situations where bringing foreign banks in, particularly when they take over banks, is likely to reduce the incidence of bank crises: when the banks take over formerly government-owned banks or when they take over owner-managed banks. Government ownership often results in sub- par average returns; owner-management often results in high variance of returns.

Ownership by foreign banks can address both problems. However, in both situations the increasing role in the banking system of foreign banks is likely to be correlated with an apparent reduction in credit outstanding.

The first situation where foreign ownership of banks may help reduce the likelihood of a recurrence of crisis occurs when the foreign banks acquire formerly government-owned banks. That is, foreign ownership of the banks impedes a government’s exploitation of its banks to keep alive failing, but politically sensitive companies, and the concomitant need 25

periodically to bail out the banks. Gros (2003) points out that permitting foreign banks to

takeover government banks can constitute a commitment and transparency device. As part

of a series of measures aimed at “limiting Leviathan”, in June 1997 the Bulgarian

government adopted a currency board, which removed the government’s control over

monetary policy and stabilized the currency (Koford 2000; Appendix 2). The government

also pushed forward privatization of the banking system to limit its own power to use the

banking system to support bankrupt firms. By privatizing the banks, the government

removed them from its own direct control. This was also the motivation in the cases of the

Czech Republic (Appendix 3) and Hungary (Appendix 4).

La Porta et al., (2002) found that higher government ownership of banks in 1970 was associated with slower subsequent financial development and slower growth in per capita income and productivity. The argued that their evidence supported “political” theories that government ownership of banks politicizes resource allocation and reduces inefficiency. In a wide-ranging study, Dinç (2002) found that government-owned banks increased their lending in election years. Restructured and overdue loans also increased. Despite the claim that government-owned banks fund projects private banks cannot finance, government- owned banks tended to fund the government by holding more government debt than private banks did. Both government-owned and private banks held a similar share of loans to assets on average across the electoral cycle.

As Sevic (2001) points out, in small states the relationship between citizens, civil servants, and politicians is often close. The government-owned banks are thus vulnerable to problems of cronyism, support buying, and corruption. In a number of cases described in

the Appendices (e.g., Jamaica, the Pacific Islands, and Tanzania) the solution has been the

privatization of the government-owned bank after its restructuring. 26

The second reform situation involves owner-managed banks. In East Asia, family

control of corporations is ubiquitous. The percentage of total market capitalization that the

top 10 families control ranges from 2 per cent in Japan (which is almost an order of

magnitude lower than the 18 per cent in Taiwan, the next lowest country) to 58 per cent in

Indonesia (Claessens et al., 1999a&b). The figure for Singapore is 27 per cent, almost

identical to that in Korea and not much greater than the 25 per cent in Malaysia. In the

Philippines, the figure is 53 per cent; one family, the Ayalas, alone accounts for 17 per cent.

Owner-managed firms have different governance problems from those of publicly

owned firms. A key strength of owner-managed firms is the absence of principal-agent

problems, and there is some evidence that owner-managed firms have a better performance

than publicly owned firms do (Tan et al. 2001). However, owner-managed firms possibly

also have a higher variance in their performance because the owner-manager does not have

to persuade others, such as a board of directors, of his strategy. No one can say no, either to

good ideas or to bad ones. In addition, often the owner-managed bank is part of an

economic group. In these cases one can expect that top management will overrule cautious

credit officers should the officers be reluctant to lend to a related company.

It is therefore not surprising that Laeven (1999) found that family-owned and

company-owned banks were among the most risky banks, whereas foreign-owned banks took little risk relative to other banks in East Asia. Furthermore, the banks requiring restructuring after the 1997 Asian crisis were mostly family-owned or company-owned and tended to have had excessive credit growth. They were rarely foreign-owned.

When these owner-managed banks got into trouble frequently the only potential rescuers were foreign banks as even the sounder host country banks were still weakened by the crisis. For instance, related and crony lending was clearly a problem in Mexico to the 27 point that one could describe the situation as one of looting (La Porta et al., 2003). It was also a problem in Thailand (Wiwattanakantang et al. 2002). One consequence of the crisis was that several of the former owner-managed banks became foreign-owned and many others had to undertake governance reforms in order to be able to draw foreign (minority) capital (Appendix 12). Related and crony lending was also a problem in Indonesia. Here, however, impediments to foreign ownership meant that the government generally had to acquire the failed banks (Appendix 5).

In both the government ownership and owner-management situations, transfer of the affected banks to foreign ownership may correlate with a decline in total loans outstanding in the banking system. One reason is an artifact of the rescue itself and the other is a consequence of the change in ownership.

To make a failed bank saleable, whether it is government owned or owner-managed, the regulatory authorities generally must first strip it of its worst assets, putting them into an asset recovery unit. This was the pattern, for instance, in the privatizations of Argentine provincial banks in the mid-1990s (Clarke and Cull 2002). Any comparison of the failed bank before and after the advent of foreign ownership will show a decline in the size of the loan book, but the direction of the causality is from reduction to the change in ownership, not vice-versa. Also the comparison conceals the fact that the accounts, by not marking the loans to their salvage value, overstated the loans outstanding before transfer. Clarke et al.,

(2003) found that privatization of Argentine provincial banks was associated with a temporary reduction in credit due to cleaning of the portfolios of the banks that were being privatized. Thereafter, growth in lending by the privatized and other banks restored credit to pre-privatization levels within a few years. 28

Still, foreign ownership will automatically bring a change in lending policy, completely apart from any effect of more stringent standards. The result will be that the new owners will reduce or even stop their lending to many of the acquired bank’s traditional customers. Until the bank can develop a new customer base, something that may take time, especially when the host country economy is in recession, the size of the loan book will drop and investments in securities, such as government bonds, will increase.

There is one situation where foreign bank entry is likely to increase the likelihood of crisis even though the entry generates an expansion of credit. The problem occurs when the entry disrupts a cartel. One technology for protecting depositors is to reduce the probability of banks failing. Governments frequently accomplish this by permitting banks to cartelize.

In their insightful paper, Breton and Wintrobe (1978) explain regulation by “moral suasion” as an exchange between the authorities and the commercial banks. The authorities provide information and other services that facilitate collusion, and deter entry. Weak banks survive and strong banks live comfortably. In return, the banks comply with the goals of the authorities, including submitting to implicit taxes such as social policies that build support for the authorities.

The entry of foreign banks can undermine the exchange for at least two reasons.

First, the increase in the number and variety of participants complicates the task of establishing agreement among the banks vis-à-vis the regulators. For the foreign banks, the operations in say, Trans-Amazonia, are only a small part of their total operations whereas operations there represent almost all of the activities of Trans-Amazonian banks. The foreign banks are also likely to have a different mix of activities than local banks. Foreign banks are, therefore, likely to react in different ways to the authorities’ strictures than will

Trans-Amazonian banks. Second, the foreign banks, if they do not wish to cooperate, can 29 appeal for support to their home governments, making domestic policy issues matters of international trade and investment policy.

The net effect on lending of the entry of the foreign banks depends on the balance between two opposing effects. On the one hand, the disruption of the cartel should lead to higher rates on deposits, lower rates on loans, and increasing lending. Levine (2002) finds that countries that restrict foreign banks from entering, even if they permit domestic entry, suffer from higher net interest margins. On the other hand, narrower margins may put some host country banks at risk. If they fail this will, at least temporarily, disrupt banking relationships and may reduce lending. Still, Clarke et al. (2001) find some evidence that the net effect on credit is positive.

One last issue has to do with government-owned banks even when these have not failed. As Marichal (1997) points out, dominance of banking by government-owned banks, especially in Argentina, Brazil, Chile and Mexico, dates from the 19th century. The government favored these banks with its business and with other concessions, but at the same time assigned these banks a development or policy role. Today, this may take a variety of forms, including the maintenance of bank branches in rural areas and small towns where business would not normally justify a bank branch, lending for agriculture or other favored sectors on better-than-market terms, and sometimes, simply the provision of jobs. What has made this possible is an implicit cross-subsidy scheme. These government-owned banks have implicitly taxed the urban and competitive corporate sectors while subsidizing the policy targets. The tax has taken the form of wider than otherwise necessary spreads between deposit and lending rates. In this context, there may not be significant differences in the performance of private and government-owned banks. Sarkar et al., (1998) found that 30 in the absence of well-functioning capital markets, public and private enterprises in India’s banking industry performed comparably.

The entry of foreign banks can undermine the cross-subsidy system. The foreign banks offer better deposit rates to urban middle and upper class customers, and better lending rates to the same customers on credit cards and mortgages, and better lending rates to profitable corporations. This leaves the government-owned banks with the burden of the policy branches and loans, but with a reduced ability to fund them. The result then is that the government-owned banks appear less profitable and less competent relative to the foreign-owned banks. The government-owned banks may be less well-run, but this is then a mandated inefficiency. Ultimately, the government faces the problem that if it privatizes these apparently unprofitable banks that it owns, the result will be that the new owners will close uneconomic rural branches, call-in unprofitable loans, and initiate mass redundancies among bank employees. These consequences may impede the privatization process by creating opponents. As Clarke and Cull (2002) found in the case of the privatization of

Argentinean provincial banks, overstaffing, high levels of provincial unemployment and higher shares of public employment all correlated with a lower likelihood of privatization.

A possible positive outcome of the whole process may be an increased transparency for the costs of the government’s policies. The problem is that the benefits of the subsidies may be less quantifiable and some socially worthwhile policies such as the integration of rural areas into the modern economy may suffer. In principle, the government can initiate a system of explicit subsidies to banks, for instance, to maintain rural branches, or programs for them to act as administering agents for loan programs. However, such policies are easier to posit in the abstract than to establish in the face of political and practical difficulties.

31

4.3) Limitations

Although foreign banks can rescue ailing institutions, remove assets from direct government control, bring competition and transfer technology, they come with handicaps.

One is the liability of foreignness and another is a liability of size.

The liability of foreignness simply refers to the fact that the foreign bank is operating at a distance and in an unfamiliar environment (Hymer 1976/1960). Both of these factors are likely to degrade the quality of the decisions that the foreign bank makes, at least relative to those that host-country firms would make, because of their effect on the quality of information. This is the argument that Hymer and later Kindleberger (1969) made for why it is necessary for foreign firms to have an offsetting advantage if they are to survive the competition from domestic firms.

Of course in acquiring a host country subsidiary the parent bank acquirers local managers who have local knowledge. However, it may be instructive to think about two extreme scenarios. If the parent gives the local managers no discretion, then it effectively makes no use of their knowledge. If the foreign bank is competing with equally capable local banks the foreigner is at a disadvantage. If the parent gives the local managers complete discretion then the parent is conceding that its decisions add no value. The more capable its local managers become, the more authority the foreign parent should devolve.

Ultimately the question will arise as to the point of owning the foreign operation when shareholders could hold it directly. One can make the argument that foreign banks can bypass their need for local knowledge by using credit scoring approaches. Clarke et al.,

(2003) conjecture that it was foreign banks’ increased use of such models that enabled them to increase their lending in provinces outside of Buenos Aires. However, capable local 32 banks can use such models too, again leaving the foreign banks with no offsetting edge vis-à- vis the liability of foreignness.

The liability of size refers to the argument that large banks and small banks do different things, but that necessarily the foreign banks that acquire host country banks are larger. Berger et al. (2002) argue, on the basis of theories of incomplete contracting, that small banks may be better than large banks at processing soft information. Bank lending to small firms is one area that relies heavily on soft information in developed economies, and even more so in developing economies where the infrastructure to produce hard information is, at best, nascent. Berger et al. (2002) find that large banks are less willing than small banks to lend to firms that do not keep formal financial records and are more likely to have impersonal, shorter, and less exclusive relationships with borrowers. Similarly, using data from Texas, Brickley et al. (2003) found that small locally-owned banks have a comparative advantage over large banks within specific environments.

By necessity, the foreign banks that participate in restructuring tend to be large.

Much depends though on how much autonomy the foreign banks allow their subsidiaries to have. If the banks the foreign banks acquire act like even larger banks, it follows then that one can expect that foreign-owned banks will tend to neglect lending to small firms. If they act like autonomous entities the effect may be muted, but then the parent’s ownership is not adding value.

To address this question, Clarke et al. (2001) used bank level data for Argentina,

Chile, Colombia and Peru for the mid-1990s. They have mixed results, in part because they do not seem to have distinguished between the strategies of foreign banks or adjusted for the time since the foreign parent acquired the host country subsidiary, something that may matter if the foreign parent’s first order of business is to change the credit policy at its 33 acquisition. Berger et al. (2001), using a data set on Argentine banks, firms and loans, found evidence suggestive of the possibility that large and foreign banks may have difficulty extending relationship loans to informationally opaque small firms. The impact of foreign rescuers on lending to small and medium firms remains an open question.

If foreign banks tend to lend to informationally less opaque firms, principally foreign firms and large domestic firms, this can give rise to charges that the foreign banks “cream skim” or “cherry pick”, leaving the worst risks to the domestic banks. Barajas et al., (2000) found that foreign entry in Colombia’s banking sector improved bank behavior by enhancing operating efficiency and competition, but may have resulted in increased risk and a subsequent deterioration in loan quality, particularly among domestic banks.

However, this is precisely the result one would hope would occur. Each class of banks is lending where it has a comparative advantage. So long as the domestic banks charge appropriate risk-adjusted interest rates there is no issue.

One might then ask what about the risk to the system through deposit insurance? If foreign and domestic banks pay the same deposit insurance premium, in that case one may conjecture that domestic banks are under-paying for deposit insurance and the foreign banks are over-paying for it. Then the outcome of the combination of “cherry picking” and deposit insurance would be a transfer from foreign banks to domestic banks.

Furthermore, when foreign banks specialize in lending to good risks they are in the long-run competing with commercial paper and bond markets, and therefore specializing in the least profitable markets, leaving the more profitable markets to the domestic banks.

Lastly, the evidence for India for the 1995-96 to 2000-01 period is that foreign banks were aggressive in credit markets rather than “cherry picking” (Bhaumik and Piesse 2003).

34

5) Conclusion

What foreign banks can bring to banking systems that have undergone a crisis is two services. The most obvious service is that foreign banks can recapitalize and rehabilitate failed banks. In many cases the new foreign parent also accomplishes a skill transfer, inculcating a credit culture and risk management. The more important service that foreign banks can perform is to be instruments of macro-economic reform. By taking over formerly government or owner-managed banks the foreign banks can reduce the likelihood of crises.

Government ownership and owner-management are subject to specific weakness that ownership by foreign banks can obviate. Experience has shown that one does not have to induce foreign banks to provide these services.

As Nikolai Bukharin (1917) wrote:

“It is finance capital that appears to be the all-pervading form of capital, that form which, like nature, suffers from a horror vacui, since it rushes to fill every "vacuum," whether in a "tropical," "sub­tropical," or "polar" region, if only profits flow in sufficient quantities.”

Bukharin correctly identified that some foreign banks will react to opportunities.

However, he failed to specify a pre-condition and to identify a consequence. The pre- condition is that foreign banks require that conditions be propitious. This means that foreign banks will take over cleaned up banks, but there are few signs that the foreigners themselves are willing to assume the burdens of writing off loans. It also means that foreign banks are more likely to enter after reforms that facilitate their operations.

The consequence that Bukharin missed is that the foreign banks’ initial entry is only a phase in a succession process. Some part of the entry is normal international banking activity that will persist indefinitely, but never be of much quantitative importance. A second part is a betting on the success of the transition or restructuring process. As their bets succeed or fail, the bettors will sell their shares, either for a good price or a low one. 35

The third part, and the one that draws the most concern in many countries around the

world, is the majority ownership of large commercial banks by foreign strategic investors, that is, investors that manage the banks and who entered either in an exploratory foray that led them further or who entered to restructure a major host country bank.

However, to the degree that the strategic investors are successful, any predominance in host economies is likely to erode over time. As the banks, foreign and domestic-owned alike, become more competitive and adept, the foreign owners will no longer have a comparative advantage in general retail and commercial banking, even if they retain an absolute advantage. Thus, in time, we can expect many foreign owners, if not all, to sell their retail banking activities to domestic owners. They may retain a branch in the national financial center to conduct corporate and wholesale banking but will leave retail and general commercial banking to local banks. More importantly, the reforms that accompanied the entry of the foreign banks will free the domestic banks to compete and grow, increasing their relative importance. This succession process will decades at best, but it is ongoing. 36

Bibliography

Barajas, A. R. Steiner, and N. Salazar (2000) The impact of liberalization and foreign investment in Colombia’s financial sector. Journal of Development Economics Vol. 63, 157-196.

Barth, J.R., G. Caprio Jr., and R. Levine (2003). Bank regulation and supervision: What works best. Journal of Financial Intermediation (forthcoming). Also, NBER WP9323.

Beck, T., A. Demirgüç-Kunt, and R. Levine (2003). Bank Concentration and Crises. World Bank, unpublished paper.

Berger, A.B., L.F. Klapper and G.F. Udell (2001). The ability of banks to lend to informationally opaque small businesses. Journal of Banking and Finance Vol. 25, 2127- 2167.

Berger, A.B., N.H. Miller, M.A. Petersen, R.G. Rajan, and J.C. Stein (2002). Does Function Follow Organizational Form? Evidence from Lending Practices of Large and Small Banks. NBER Working Paper 8752.

Bhaumik, S.K. and J. Piesse (2003). Are Foreign Banks Bad for Development Even if They are Efficient? Evidence from the Indian Banking Industry. William Davidson Institute WP 619.

Bonin, J.P. and P. Wachtel (1999). Towards Market-Oriented Banking in the Economies in Transition. In M. Blejer and M. Skreb, eds., Financial Sector Transformation: Lessons from the Economies in Transition, Cambridge Univ. Press, Cambridge.

Breton, A. and R. Wintrobe (1978). A Theory of 'Moral' Suasion. Canadian Journal of Economics, Vol. 11, 210-219.

Brickley, J.A., J.S. Linck and C.W. Smith, Jr. (2003) Boundaries of the firm: evidence from the banking industry. Journal of Financial Economics Vol. 70, 351-383.

Buch, C.M. (1997). Opening Up for Foreign Banks: How Central and Eastern Europe can Benefit. Economics of Transition, Vol. 5(2), 339-366.

Bukharin, N. (1917). Imperialism and World Economy, International Publishers, New York (first published in English in 1929), Chapter 3. http://www.marxists.org/archive/bukharin/works/1917/imperial/03.htm

Calomiris, C.W., D. Klingebiel and L. Laeven (2003). A Taxonomy of Financial Crisis Restructuring Mechanisms: Cross-Country Experience and Policy Implications. In L. Laeven (ed.) Systemic Financial Distress: Containment and Resolution.

Claessens, S., A. Demirgüç-Kunt, and H. Huizinga (2001). How Does Foreign Entry Affect Domestic Banking Markets? Journal of Banking and Finance, Vol. 25, 891-911. 37

Claessens, S., S. Djankov, and L.H.P. Lang (1999a). Who Controls East Asian Corporations? World Bank Policy Research 2054.

Claessens, S., S. Djankov, and L.H.P. Lang (1999b). Who Controls East Asian Corporations – and the Implications for Legal Reform. World Bank Public Policy for the Private Sector Note 195.

Claessens, S. and J-K. Lee (2002). Foreign Banks in Low-Income Countries: Recent Developments and Impacts. World Bank, unpublished paper.

Clarke, G.R.G., J.M. Crivelli and R. Cull (2003) The Impact of Bank Privatization and Foreign Entry on Access to Credit in Argentina’s Provinces. World Bank. Unpublished paper.

Clarke, G.R.G., and R. Cull (2002) Political and Economic Determinants of the Likelihood of Privatizing Argentine Public Banks. Journal of Law and Economics Vol. 45, 165-197.

Clarke, G.R.G., R. Cull, and M. S. Martinez Peria (2001). Does Foreign Bank Penetration Reduce Access to Credit in Developing Countries? Evidence from Asking Borrowers. World Bank Policy Research WP 2716.

Dages, B.G., L. Goldberg and D. Kinney (2000). Foreign and domestic bank participation in emerging markets: Lessons from Mexico and Argentina. Federal Reserve Bank of New York Economic Policy Review September, pp. 17–36.

Demirgüç-Kunt, A., and H. Huizinga (1999). Determinants of Commercial Bank Interest Margins and Profitability: Some International Evidence. World Bank Economic Review, Vol. 13(2), 379-408.

Dinç, I.S. (2002). Politicians and Banks: Political Influences on Government-Owned Banks in Emerging Countries. Univ. of Michigan, unpublished paper.

Dopico, L.G., and J.A. Wilcox (2002). Openness, Profit Opportunities and Foreign Banking. Journal of Financial Markets, Institutions, and Money, Vol. 12(4-5), 299-320.

Du, J. (2003). Why do multinational enterprises borrow from local banks? Economics Letters Vol. 78, 287-291.

Engwall, L., R. Marquardt, T. Pedersen, and A.E. Tschoegl (2001). Foreign Bank Penetration of Newly Opened Markets in the Nordic Countries. Journal of International Financial Markets, Institutions, and Money, Vol. 11(1), 53-63.

Fane, G. and R.H. McLeod (2002). Banking Collapse and Restructuring in Indonesia, 1997- 2001. Cato Journal, Vol. 22 (2), 227-295.

Goldsmith, R. W. (1969). Financial structure and development. New Haven, Yale University Press.

Gros, D. (2003). Who Needs Foreign Banks? CESifo WP 998. 38

Gruben, W.C. and R.P. McComb (2003). Privatization, Competition, and Super Competition in the Mexican Commercial Banking System. Journal of Banking and Finance, Vol. 27, 229-249.

Guillén, M. and A.E. Tschoegl (2000). The Internationalization of Retail Banking: The Case of the Spanish Banks in Latin America. Transnational Corporations Vol. 9(3), 63-97.

Harrington, D.G. and T.C. Lawton (2003). The Internationalization of Allied Irish Banks. ECCH Case 303-015-1. Hasan, I. and K. Marton (2003). Development and efficiency of the banking sector in a transitional economy: Hungarian experience. Journal of Banking and Finance Vol.27, 2249-2271..

Harvey, C. (1991). On the Perverse Effects of Financial Sector Reform in Anglophone Africa. South African Journal of Economics, Vol. 59(3), 258-286.

Heinkel R. L., and M.D. Levi (1992) The structure of international banking. Journal of International Money and Finance Vol. 16, 251-72.

Hewison, K. (2001). Pathways to Recovery: Bankers, Business and Nationalism in Thailand. City University of Hong Kong, Southeast Asia Research Centre, Working Paper 1.

Honohan, P. (2003). Recapitalizing Banking Systems: Fiscal, Monetary and Incentive Implications. In L. Laeven (ed.) Systemic Financial Distress: Containment and Resolution.

Huang, Y., and D. Wi (2003) A Tale of Two Provinces: The Institutional Environment and Foreign Ownership in China. MIT, unpublished paper.

Hymer, S.H. (1976). The International Operation of National Firms: A Study of Direct Investment (Boston, Mass.: MIT; 1960 thesis).

Ignatiev, P. (1997). Development of the Bulgarian Banking System between 1992 and 1996. Bulgarian National Bank, Monthly Bulletin, 27-31, July.

Kindleberger, C.P. (1969). American Business Abroad. New Haven, Yale University Press.

Király, J., B. Májer, L. Mátyás, B. Öcsi, A. Sugár, and E. Várhegyi (2000). Experience with Internationalization of Financial Sector Providers – Case Study: Hungary. In: S Claessens and M. Jansen, eds., The Internationalization of Financial Services: Issues and Lessons for Developing Countries, Kluwer Academic Press, Dordrecht.

Kirkpatrick, C. and D. Tennant (2002). Responding to Financial Crisis: The Case of Jamaica. World Development, Vol. 30(11), 1933-1950.

Koford, K. (2000). Citizen Restraints on “Leviathan” Government: Transition Politics in Bulgaria. European Journal of Political Economy, Vol. 16, 307-338.

Koford, K. and A.E. Tschoegl (2002). Foreign Banks in Bulgaria, 1875-2002. Univ. of Delaware WP02-06, William Davidson Institute, Working Paper 537. 39

Laeven, L. (1999). Risk and Efficiency in East Asian Banks. World Bank Policy Research Working Paper 2255.

La Porta, R., F. Lopez-de-Silanes, and A. Shleifer (2002). Government Ownership of Banks. Journal of Finance, Vol. 57, 265-301.

La Porta, R., F. Lopez-de-Silanes, and G. Zamarripa (2003). Related Lending. Quarterly Journal of Economics Vol. 118 (1), 231-267.

Lensink, R. and N. Hermes (2003) The short-term effects of foreign bank entry on doestic bank behaviour: Does economic development matter? Journal of Banking and Finance (Forthcoming).

Levine, R. (2002). Denying Foreign Bank Entry: Implications for Bank Interest Margins. Univ. of Minnesota, unpublished paper.

Lindblom, T. (1994). Credit Losses in Nordic Banks. In Jack Revell, ed. The Changing Face of European Banks and Securities Markets, St. Martins Press, London.

Lwiza, D.R.B. and S. Nwankwo (2002). Market-Driven Transformation of the Banking Sector in Tanzania. International Journal of Bank Marketing, Vol. 20(1), 38-49.

Majnoni, G., R. Shankar, and E. Varhegyi (2003) The Dynamics of Foreign Bank Ownership: Evidence from Hungary. World Bank Policy Research WP 3114.

Marichal, C. (1997). Nation Building and the Origins of Banking in Latin America, 1850- 1930. In A. Teichova, G. Kurgan-van Hentenryk, and D. Ziegler, eds. Banking, Trade and Industry: Europe, America and Asia from the thirteenth to the twentieth century, Cambridge Univ. Press, Cambridge.

Miller, J.B. and S. Petranov (1996). Banking in the Bulgarian Economy .Sofia, Bulgarian National Bank.

Miller, J.B. and S. Petranov (2001). The Financial System in the Bulgarian Economy (Sofia: Bulgarian National Bank).

Minushkin, S. (2002). Banqueros and Bolseros: Structural Change and Financial Market Liberalization in Mexico. Journal of Latin American Studies Vol. 34, 915-944.

Moffett, M.H., A. Stonehill, and M. Balling (1989). Multinational Bank Strategy in Theory and Practice: Towards the EC Internal Market and 1992. Oregon State University, unpublished paper.

Peek, J. and E.S. Rosengren, (2000). Collateral Damage: Effects of the Japanese Bank Crisis on Real Activity in the United States. American Economic Review, Vol. 90(1), 30-45.

Phylaktis, K. (1988). “Banking in a British colony: Cyprus, 1878-1959”, Business History, 30 (4), 416-431. 40

Reynoso, A. (2002). Can Subsidiaries of Foreign Banks Contribute to the Stability of the Forex Market in Emerging Economies? A Look at Some Evidence from the Mexican… NBER Working Paper 8864.

Sarkar, J., S. Sarkar, and S.K. Bhaumik (1998) Does Ownership Matter?—Evidence from the Indian Banking Industry. Journal of Comparative Economics Vol. 26, 262-281.

Šević, Ž. (2000). Banking Reform in South East European Transitional Economies: An Overview. MOCT-MOST Vol. 3-4, 271-283.

Sevic, V. (2001). Politico-Administrative Relationships in Small States. Bank of Valetta Review, Vol. 23(Spring), 63-76.

Siwatibau, S. 1996. The National Bank of Fiji. Pacific Economic Bulletin Vol. 11(1), 76-84.

Stennet, R.R; P.M. Batchelor, and C. Foga (1999) Stabilisation and the Jamaican Commercial Banking Sector (1991-1997) Social and Economic Studies 48 (1&2,).

Tan, R.S.K., P.L. Chng and T.W. Tan (2001). CEO ownership and Firm Value. Asia Pacific Journal of Management, Vol. 18(3), 355-371.

Tang, H., E. Zoli, and I. Klytchnikova (2000). Banking Crises in Transition Countries: Fiscal Costs and Related Issues. World Bank, Country Economics Dept., Paper 2484.

Triner, G.D. (1996). The Formation of Modern Brazilian Banking, 1906-1930: Opportunities and Constraints Presented by the Public and Private Sectors. Journal of Latin American Studies Vol. 28(1), 49-74.

Triner, G.D. (2002) British Banking in Brazil during the First Republic. Rutgers Univ., unpublished paper.

Tschoegl, A.E. (1981). The Regulation of Foreign Banks: Policy Formation in Countries Outside the United States. Salomon Brothers Center for the Study of Financial Institutions, Monograph Series in Finance and Economics, Monograph No. 1981-2.

Tschoegl, A.E. (1988). Foreign Banks in Japan. Bank of Japan, Institute for Monetary and Economic Studies, Monetary and Economic Studies Vol. 6(1), 93-118.

Tschoegl, A.E. (2001). Foreign banks in the United States since World War II: A useful fringe. In Geoffrey Jones and Lina Gálvez-Muñoz, eds. Managing Foreign Business in the US (London: Routledge), 149-168.

Tschoegl, A.E. (2002a). Entry and Survival: The case of foreign banks in Norway. Scandinavian Journal of Management, Vol. 18(2), 131-153.

Tschoegl, A.E. (2002b). FDI and Internationalization: Evidence from US Subsidiaries of Foreign Banks. Journal of International Business Studies, Vol. 33(4), 805-815. 41

Tschoegl, A.E. (2002c). Foreign Banks in Saudi Arabia: A Brief History. Transnational Corporations, Vol. 11(3), 123-154.

Tschoegl, A.E. (2003a). The California Subsidiaries of Japanese Banks: A Genealogical History. Wharton School, unpublished paper.

Tschoegl, A.E. (2003b) “The world’s local bank”: HSBC’s expansion in the US, Canada and Mexico. Wharton School, unpublished paper.

Tschoegl, A.E. (2004) Who Owns the Major US Subsidiaries of Foreign Banks? A Note. Journal of International Financial Markets, Institutions & Money (Forthcoming).

Unite, A.A. and M.J. Sullivan (2002). The effect of foreign entry and ownership structure on the Philippine domestic banking market. Journal of Banking and Finance Vol. 27, 2249- 2271.

Weill, L. (2003) Banking efficiency in transition economies: The role of foreign ownership. Economics of Transition Vol. 11(3), 569-592.

Williams, B. (1996). Determinants of the Performance of Japanese Financial Institutions in Australia 1987-1992. Applied Economics, Vol. 28(9), 1153-65.

Wiwattanakantang, Y., R. Kali, and C. Charumilind (2002). Crony Lending: Thailand before the Financial Crisis. Center for Economic Institutions, Institute of Economic Research, Hitotsubashi University.

Worrell, D., D. Cherebin, and T. Polius-Mounsey (2001). Financial System Soundness in the Caribbean: An Initial Assessment. IMF Working Paper 01-123. 42

Table 1a: Allied Irish Banks Poland: In the late 1980s AIB had a surplus of senior managers so it lent them out as consultants working in the transition economies, particularly the Czech Republic, Hungary and Poland. In 1993, at the request of the IBRD and the EBRD, AIB agreed to a twinning arrangement with Wielkopolski Bank Kredytowy (WBK). In 1995 AIB took a 16.3 per cent stake in WBK. In 1996 it acquired a further 20 per cent from the Polish government. Then in 1997 AIB acquired shares in WBK from the EBRD, bringing its ownership to 60 per cent. In 1999 AIB reached an agreement with the Polish State Treasury to acquire an 80 per cent shareholding in Bank Zachodni. Then in 2000 AIB merged Wielkopolski Bank Kredytowy and Bank Zachodni to form Bank Zachodni WBK. AIB currently owns 70.50 per cent of BZ WBK and the remaining shares are widely held. Both WBK and Zachodni were regional retail banks rather than banks to the corporate sector. Singapore: In 1999 AIB entered into a strategic alliance with Keppel Tat Lee Bank (KTL) in Singapore. AIB acquired one per cent of the shares of KTL and an option to acquire a further 25 per cent. However, in 2001 Oversea Chinese Banking Corporation, also of Singapore, made an offer to acquire KTL. AIB chose to sell out to OCBC, at a notable profit. Note: Having sold its AllFirst subsidiary in the US (Tschoegl 2002b), AIB has no retail banking operation outside Ireland and the UK other than Bank Zachodni WBK. Source: Harrington and Lawton (2003)

Table 1b: Prager Handelsbank 1989 Four banks established Prager Handelsbank (PH): Ceskoslovenska Obchodni Banka (CSOB; 45%), DG Bank (30%), Zivnostenska Banka (10%), Raiffeisen Zentralbank Oesterreich (Wien) (10%) and BHF Bank (5%). The plan was that PH would specialize in financing foreign trade transactions, especially between Germany, Czechoslovakia and Austria. 1990 PH opened its head office in Frankfurt. 1991 PH opened a rep office in Prague but for legal reasons could not open a branch. 1992 CSOB (55%) and DG Bank (35%) acquired the shares of Zivnostenska banka and BHF-Bank. BHF-Bank had acquired a 40% stake in Zivnostenska banka at the beginning of thes year. 1997 CSOB bought out its partners in PH. 2000 CSOB sold PH to Belgium’s KBC Bank. KBC had acquired 66% of CSOB in May 1999. At the end of the 1999 it obtained another 16.66% during the sale of a stake in CSOB held by the National Bank of Slovakia. KBC integrated PH into KBC Deutschland and the name disappeared. Source: News reports.

43

Table 2: European banks’ patterns of acquisition in Central and Eastern Europe

Unicredito Erste Bank KBC Société Générale (Italy) (Austria) (Belgium) (France) Poland Bank Pekao Kredyt Bank

Czech Zivnostenska Ceska sporitelna CSOB Komercni banka Republic Banka

Slovakia Unibanka Slovenska CSOB Komercni banka sporitelna Bratislava

Slovenia Nova SKB Banka Ljubljanska Banka1

Croatia Zagrebacka Erste & Banka Steiermärkische Bank (ex- Bjelovarska banka, Trgovacka banka, Cakovecka banka, and Rijecka banka)

Serbia Société Générale Yougoslav Bank

Hungary Erste Bank K&H Bank Hungary (ex-Mezöbank) Postabank

Romania Unicredit Banque Romania Roumaine de Développement

Bulgaria Bulbank SG Expressbank

Note: 1) Associated bank. Sources: Bank websites

44

Appendix 1: Argentina

Argentina is a special case. In late 2001 the government of former President

Fernando de la Rua defaulted on most of Argentina’s US$141bn in public external debt, devalued the peso (breaking the currency board parity of 1 peso = US$1), froze deposits, and introduced capital controls. It also engaged in asymmetric pesification, mandating the conversion to pesos of bank accounts dominated in US dollars, but with a higher rate for deposits than for loans.

Before the government precipitated the full crisis foreign banks participated in several rescues. In 1999 Banca Nazionale del Lavoro, Banco Bisel, Banco Macro Misiones,

Banco Sudameris Argentina, Banco Supervielle and Banco del Suquia acquired the branches of the provincial bank Banco Israelita de Cordoba. The government had partially suspended

Israelita due to a significant outflow of deposits.

In early 2001, the government sold off MBK Mercobank in pieces because its shareholders refused to commit to a US$10mn capital increase to continue operations. The leading shareholder, Chilean investment group Inversora Norte-Sur, simply walked away from its investment. Crédit Agricole, another shareholder, also refused to subscribe to the increase. Ten banks (half of them foreign-owned) split up Mercobank’s 47 branches amongst themselves.

Later that year Galicia, Sudameris (see below), Credicoop, San Luis and Nuevo

Industrial de Azul took over management of Banco Balcarce’s eight branches and 14 service centers. The Central Bank had assumed control of Balcarce in response to accounting irregularities and its inadequate capitalization.

In late 2001 concerns about the bank’s liquidity led the central bank to suspend

Scotiabank Quilmes's operations. Scotiabank (formerly Bank of Nova Scotia) refused to 45 pump in more capital and instead abandoned Scotiabank Quilmes, writing off C$540mn in the process. Scotiabank had acquired the 91-branch bank in 1997. The government transferred Quilmes to Banco Comafi and Banco Bansud, which appeared to have roughly split it between them, and which took over all the liabilities, not just the deposits.

Scotiabank did pay severance pay to the 500 (of 1700 employees) who lost their jobs. To help depositors, it also did not claim some parent assets still with Quilmes. Even though

Scotiabank had no further legal liability, it has offered C$37mn to pay 20 cents on the dollar to holders of Quilmes’ medium term notes and other debt. Scotiabank apparently made the offer as a sign of good faith and out of a concern for its reputation in debt markets.

Crédit Agricole exercised the option implicit in limited liability and walked away from its three subsidiaries in 2002. It simply refused to send the around 200mn pesos

(US$61.5mn) they would require to stay afloat. The three subsidiaries—Banco Bisel, Banco del Suquia and Banco de Entre Rios— have 353 branches that serve farmers in the provinces. The government-owned Banco de la Nación Argentina, the largest bank in the country, has temporarily stepped in to take over the three to keep them running.

In 2003 IntesaBCI sold Banco Sudameris Argentina (BSA), to Banco Patagonia, a provincial bank. In terms of book capital, BSA is about four times larger than Patagonia.

Intesa had acquired the Sudameris chain of banks in Latin America in 1999 with its acquisition of Banca Commerciale Italiana and had been seeking to dispose of the various subsidiaries ever since. BSA, which had received 400 million pesos (US$110mn) in loans from the Central Bank in 2002, has pledged to boost capital by 100mn pesos to close the transaction. IntesaBCI is taking a 20 per cent stake in the new entity and contributing an unspecified amount of cash to provide working capital for the bank. 46

Appendix 2: Bulgaria

The regime of Todor Zhivkov ended on 10 November 1989, just one day after the fall of the Berlin Wall. However, Bulgaria’s revolution was essentially a coup within the

Communist Party. The new government made only those changes it was forced to make, and initially barely accepted the entry of foreign banks. Despite the phased liberalization of entry and accumulation of foreign entrants, by end-June 1997 the foreign share of the assets of the banking system was only 4.9 per cent.

In early 1996 problems with loans to government companies led to a bank panic that in turn led to a currency crisis and a brief hyperinflation. In early 1997 the government announced that it would institute a currency board system and other reforms. Subsequently the crisis subsided. In all, the period from 1991 through the twin banking and currency crises caused Bulgaria a loss of about 42 per cent of GDP (Tang et al., 2000).

The new government began privatizing the banking system. It favored the entry of foreign banks because these could draw on their parents for liquidity when the currency board system limited the BNB’s ability to act as a lender of last resort (Šević 2000).

In May 2003 the government completed the privatization process when it sold 100 per cent of DSK, the state savings bank, to OTP (Országos Takarékpénztár és Kereskedelmi

Bank), itself the former Hungarian state savings bank. DSK accounts for about 14 per cent of assets in the Bulgarian banking system so now foreign owners account for about 86 per cent of assets in the banking system. Although the government still owns one small bank,the government has announced that it will close Bulgaria’s Bank Consolidation

Company at the end of 2003. 47

Table: Foreign-owned banks have come to dominate the Bulgarian banking system Ownership 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

State Number 72 69 25 15 12 7 6 5 7 3 3 2 Savings 1 1 1 1 1 1 1 1 Share (%) 98.596.8 93.6 84.4 76.9 82.5 51.117.6 18.1 14.2

Bulgarian Number 6 11 15 23 22 13 13 11 7 9 10 10 private Share (%) 1.5 3.2 6.4 15.6 22.4 15.2 6.9 9.9 10.3 11.2

Foreign Subsidiaries 0 3 5 9 10 13 15 16 16 Branches 2 4 4 5 7 7 8 7 6 Share (%) 0.1 0.6 2.3 25 42 73 72 74

All banks Number 79 81 41 41 42 30 34 34 34 35 35 34 Assets (BL Bn) 463 582 810 1072 1,089 3,301 8,076 7,589 8,223 9,774 11,90814,558 H2 (a) 0.38 0.33 0.30 0.15 0.11 0.19 0.12 0.11 0.120.11 0.09 0.08 Notes: (a) H2 is the Hirschman-Herfindahl index. H2 is the sum of the squares of each bank’s share of the total assets in the banking system. H2 ranges between 1, which represents a situation of monopoly, and 0, which would represent complete dispersion. Source: BNB; Ignatiev (1997); Miller and Petranov (1996); Miller and Petranov (2001). 48

Appendix 3: Czech Republic

In the euphoria that followed the "Velvet Revolution" the Czechoslovak government was

notoriously lax in its licensing of banks. After 1989 the number of Czech-controlled non-

government banks peaked at 28. Even so, Československá Obchodní Banka (CSOB), Komerční

Banka (KB), Investiçní a Poštovnia Banka (IPB), and Česká Spořitelna (CS) held about 80 per cent

of all assets. Established in 1964, CSOB had been the foreign trade bank. The government had

created KB (Commercial Bank) and IPB (Investment Bank) out of the activities of the former

Czechoslovak State Bank, which had functioned as Central bank and universal bank. CS was the former state savings bank.

The government was also reluctant to permit strategic investors, especially foreign ones.

(This is part of a long-seated distrust. In 1920, after the dissolution of the Hapsburg Empire

Czechoslovakia passed a Nostrification Law that required firms with assets in Czechoslovakia to incorporate there.) The government nominally privatized the major banks by selling just under 50 per cent of each in voucher privatizations. The remaining shares remained with entities such as the

National Pension Fund or the Ministry of Finance.

In 1992, German BHF-BANK took up a 40 per cent stake in Živnostenská Banka (ZB).

(The IFC acquired 12 per cent and private individuals and Czech investment funds acquired the remaining 48 per cent.) ZB, established in 1868, was the country’s oldest bank and once its largest.

In 1998, Bankgesellschaft Berlin took over BHF-BANK's stake and built up its ownership to 85 per cent. In 2001 Bankgesellschaft Berlin sold its stake to Italy’s Unicredito, which now plans to buy all the remaining shares.

In the mid-1990s many Czech banks failed for reasons that ranged from inexperience to theft. Most of these were those set up after the end of Communism. The government encouraged

CSOB to acquire some and others were liquidated.

49

Among the transition economies, the Czech Republic was late to accept foreign control.

Once it bowed to necessity in 1998, things moved quickly. By the end of 1999 foreign institutions controlled almost 50 per cent of total bank assets. That percentage may have reached more than 90 per cent with the privatization of Komercni Banka (Mathieson and Roldos 2001).

In 1998, Nomura Principal Investments acquired 46 per cent of IPB from the government.

In 2000 the bank verged on bankruptcy and the government seized it, three days later selling it to

CSOB (see below) with indemnities against losses. The government and Nomura are seeking arbitration, the Czech government for the losses it sustained on IPB, arguing that Nomura had undertaken to strengthen the bank but had failed to do so, and Nomura arguing that the government had failed to protect its investment.

In 1998 the government sold CS to Erste Bank, an Austrian savings bank holding company.

In 1999 the government sold CSOB to KBC, a leading Belgian Bank. After having bailed out KB twice, in 2001 the National Property Fund sold a 60 per cent stake to France’s Société

Générale. As Table 2 (above) shows, Unicredito, Erste Bank, KBC and Société Générale have all been active in the pursuit of acquisitions in the region.

50

Table: Foreign controlled banks now account for the majority of banks in the Czech Republic

Czech-controlled Foreign controlled TOTAL Unlicensed banks3 State State- Czech- Under financial owned controlled conservator institutions banks banks -ship Banks Branches 19901 4 1 0 0 0 0 5 x 19902 4 1 4 0 0 0 9 x 19912 4 1 15 0 4 0 24 x 19922 1 4 21 0 9 2 37 x 19932 1 4 28 1 12 6 52 x 19942 1 4 28 1 13 8 55 2 19952 1 6 25 0 13 10 55 5 19962 1 6 18 5 14 9 53 7 19972 1 6 15 4 15 9 50 11 19982 1 5 14 0 15 10 45 18 19992 1 4 10 0 17 10 42 21 20002 1 4 8 1 16 10 40 23 20012 0 3 8 1 16 10 38 25 20022 0 2 9 0 17 9 37 27 Notes: 1) 1 January; 2) 30/31 December; 3) Banks in liquidation, bankruptcy, non-commencement of operation, no longer operating due to merger, or transformed into a non-bank entity. Source: Česká Národni Banka

51

Appendix 4: Hungary

In 1975 Austria’s Creditanstalt Bankverein opened a representative office in Budapest to

become the first foreign bank to have a presence in Hungary in the post-WWII era. The first

foreign-owned bank incorporated in Hungary in the post-WWII era was Central-European

International Bank (CIB). CIB was a consortium bank founded in 1979 as an offshore bank to deal

primarily with international trade; it kept its books in US$ and was exempt from exchange controls.

The largest single owner was the Hungarian Central Bank (34 per cent). In 1986-87, the Hungarian

government authorized foreign banks to enter via foreign majority-owned joint ventures with

Hungarian banks. The great influx of foreign banks began in 1990. The government liberalized the

entry rules further and ING Bank became the first foreign bank to establish a wholly owned

subsidiary. Now most of the de novo entry is via wholly owned subsidiaries and the foreign owners

have bought out the Hungarian partners in the early joint ventures.

The banking crisis of the early 1990s left the government-owned banks and several private

Hungarian banks in desperate straits. The government decided not to rescue the banks through an

infusion of capital, recognizing that as long as the government owned the banks in whole or in part,

borrowers and managers would expect a bail-out, which in turn would lead to problems of moral

hazard. Having decided on privatization, the government sought strategic investors rather than

dispersed ownership even though the only possible strategic investors were foreign firms as

Hungarian non-financial firms lacked the capital and the skills. Also, the government recognized

that conflicts of interest could arise when a bank’s owners were also its largest customers.

The government’s decision to permit foreign banks to acquire Hungarian private banks and

to privatize the banking system led both to a deepening of entry and further entry (Bonin and

Wachtel 1999). The first privatization, in 1994, was that of Magyar Külkereskedelmi Bank (MKB; the foreign trade bank). The second major privatization was that of OTP, though here the

52

government preferred to disperse ownership rather than seek a strategic investor. It privatized in tranches in a process that has resulted in substantial aggregate foreign ownership though the government has apparently retained one “golden share.” Even the Hungarians were uncomfortable with foreign control of the key savings institution (Bonin and Wachtel 1999). Still, numerous other privatizations to foreign owners followed.

In 1997, Hungary agreed to OECD demands that it permit foreign banks to open branches of the parent (Király et al. 2000). Although Hungary nominally acceded to OECD pressure for further liberalization of entry, in practice the government vitiated the cost advantage of a branch vis-

à-vis a subsidiary and hence did not materially ease entry.

In 2001, Belgian KHB and Dutch ABN-AMRO Bank merged their subsidiaries, creating the

second-largest bank in the country, one that is almost as large as OTP. Earlier, HVB and Bank

Austria merged their subsidiaries to create the fifth largest bank. The full privatization of Postabank

ran into difficulties because of extensive losses at the bank, though recently (Sep 2003) the

government announced that Austria’s Erste Bank has won the bidding.

In a study of foreign bank ownership in Hungary, Majnoni et al., (2003) found that greenfield

banks clearly outpaced other foreign-owned banks in terms of profitability. However,

improvements in profitability derived from the introduction of new products and a broader array of

financial services rather than from higher intermediation margins. Hasan and Marton (2003)

examined the Hungarian banking sector during the transitional process. They found that flexible

approaches to privatization, and liberal policies towards foreign banks' involvement with the

domestic institutions helped to build a relatively stable and increasingly efficient banking system.

Furthermore, foreign banks and banks with higher foreign bank ownership involvement were

associated with lower inefficiency.

53

Table: In Hungary, foreign banks account for 8 of the top 10 banks and 65% of the assets in the banking system Rank Bank Owner Assets (1) Branches (2) (1)/(2) (Ft Bn) (%) (#) (%) (Ft Bn) 1 OTP1 2,393 30 430 42 5.6 2 K+H KBC Bank 1,196 15 163 16 7.3 3 MKB Bayerische 964 12 30 3 32.1 Landesbank 4 CIB IntesaBCI 831 10 45 4 18.5 5 HVB Bank Austria 579 7 38 4 15.2 6 Raiffeisen Bank Raiffeisen Bank 554 7 43 4 12.9 7 Postabank2 Hungarian state 407 5 114 11 3.4 8 Erste Bank Erste Bank 393 5 79 8 5.0 9 Budapest Bank GE Capital 332 4 56 6 5.9 10 Citibank Citibank 322 4 19 2 16.9 Total 7,971 100 1017 100 Notes: 1) OTP is about 30-40 per cent foreign owned, but the shares are widely held. 2) In Sep 2003 Erste Bank won the bidding for Postabank. Source: Financial Times

54

Appendix 5: Indonesia

From 1969 to the financial sector deregulation in 1988, the Indonesian government issued

no licenses for branches of foreign banks. Starting in 1988 the government permitted foreign banks to form joint-venture subsidiaries provided that 1) the foreign bank was a major bank; 2) that the home country offered reciprocity to Indonesian banks; and 3) that the Indonesia partner had at least

15 per cent of the equity. Subsequently foreign banks established a large number of these joint ventures. The government also permitted foreign banks to establish up to one branch in each of the six largest cities.

The 1988 liberalization also resulted in a rush by domestic private parties to establish banks.

As a result, Indonesia entered the crisis with 160 private domestic commercial banks alone, compared with 30-40 banks in Malaysia and 15 in Thailand. Even so, government-owned banks accounted for 34 per cent of banking system assets and 30 per cent of deposits. By contrast, foreign

JVs and branches accounted for 10 and 3 per cent.

The Asian Crisis resulted in widespread bank failure. Of the largest banks, the seven original state banks and the 10 largest formerly private banks all failed. Some continue to operate under their original names but only after receiving a government bailout. The net cost to the government of the bailout may be on the order of 40 per cent of GDP (Fane and McLeod 2002).

Since the crisis, the government has permitted foreign banks to convert joint ventures to wholly owned subsidiaries, and has started to move towards permitting foreign banks to acquire domestic banks. However, in 1999, Standard Chartered Bank of the UK called off its agreement to buy a stake in Bank Bali after a dispute with the staff of the bank. The deal had already run into difficulties earlier in the year when due diligence investigation revealed that Bank Bali was involved in the transfer of funds (later reversed) to a business connected to the then ruling Golkar Party.

Standard Chartered made a second attempt to buy an Indonesian bank but in May 2002 the

55 government sold a 51 per cent stake in Bank Central Asia (BCA) to Farallon Capital Management.

Farallon paid Indonesia $541 million for BCA, an amount less than Standard Chartered’s bid, but undertook not to cut staff numbers for two years. Farallon, a San Francisco-based hedge fund, was working with the Hartono family, which owns cigarette maker PT Djarum, Indonesia's third-largest cigarette maker, who reportedly would own 10 per cent of BCA. The deal means that the majority of the shares of BCA would remain effectively in Indonesian hands, allaying nationalist concerns.

Later in 2002 Indonesia's parliament finally approved IBRA’s sale of 51 per cent of Bank Niaga to

Commerce Asset Holdings of Malaysia, the owner of Bumiputra-Commerce Bank. In 2003 Asia

Finance Indonesia, a consortium made up of Singapore's state investment arm, Temasek Holdings, and Germany's Deutsche Bank bought 51 per cent of Bank Danamon from IBRA. Temasek owns about 34 per cent of Development Bank of Singapore (DBS), the largest commercial bank in

Singapore, but it is the owner of Danamon, not DBS. The Indonesian government has been slower to sell banks than the Thai or Korean governments, and even more reluctant to permit foreigners to acquire the banks it was selling.

Table: State-owned banks dominate in Indonesia and foreign-owned banks remain marginal Regional Foreign and State-owned Private Development joint-venture Banks 1997 7 160 27 44 2000 5 83 26 39 Branches 1997 1,748 5,133 776 90 2000 1,734 3,777 798 71 Assets 1997 152.6 237.9 12.7 44.2 34% 53% 3% 10% 2000 458.7 331.9 23.2 113.9 49% 36% 3% 12% Deposits 1997 68.6 146 8 6.9 30% 64% 3% 3% 2000 237.7 228 17.6 25.2 47% 45% 3% 5% Source: Bank Indonesia

56

Appendix 6: Jamaica

As Worrell et al. (2001) point out, Jamaica is the only Caribbean country to have suffered a major financial crisis in the 1990s. Rapid liberalization in the financial sector during the late 1980’s and early 1990’s resulted in a proliferation of both commercial and merchant banks. The government also imposed especially high reserve requirements on banks both absolutely and relative to other financial institutions. Lastly, tight monetary policy followed a period of high inflation.

However, mismanagement of the banks, especially connected lending was also a major source of problems (Stennet et al., 1999).

The Jamaican financial crisis was severe. As the problems in the banking sector grew, depositors engaged in a flight to quality, withdrawing their savings from what they perceived to be weak institutions, mainly indigenous ones, and depositing them in the branches of foreign banks

(Kirkpatrick and Tennant 2002). Towards the end of 1996, the government adopted a cohesive approach to resolving the problems in the financial sector. In January 1997 it established the

Financial Sector Adjustment Company (FINSAC), which intervened in five of the island’s nine banks. The most important intervention concerned National Commercial Bank (NCB), the largest bank in Jamaica. The other four intervened banks were smaller, but together with NCB, the interventions in total accounted for over half of the banking system’s assets. Almost the entire remainder of the assets was in the hands of foreign banks. The FINSAC bailout has left Jamaica with a high debt burden of approximately 130 per cent of GDP in 2001.

National Commercial Bank traces its origins to 1837 when the newly-formed Colonial Bank established a branch in the country. In 1925 Colonial Bank merged with two other banks to become

Barclays Bank (Dominion, Colonial and Overseas). In 1976 the Jamaican government nationalized the bank and changed its name to National Commercial Bank Jamaica. In 1986 the government privatized the bank. In 1993 NCB merged with Mutual Security Bank. In 1997 and 1998, FINSAC

57 provided assistance by taking over NCB’s non-performing loans. This was not enough. FINSAC reorganized NCB and in the process acquired 76 per cent of the shares. Early in 2002, AIC,

Canada's largest privately held mutual fund company and the 11th largest mutual fund company overall, acquired FINSAC’s shares. Michael Lee-Chin, a Jamaican-Canadian, is Chairman and CEO of AIC and owns 90 per cent of the company; bank staff members own the remaining 10 per cent.

Mr. Lee-Chin's family operates the Jamaican supermarket chain Super Plus.

RBTT is the surviving entity of the four smaller banks that FINSAC amalgamated: Citizens

Bank, Eagle Commercial Bank, Island Victoria Bank and Worker Savings and Loan Bank, and their allied merchant banks and building societies. In 1999, FINSAC injected new capital into the bank, which changed its name to Union Bank of Jamaica (UBJ). The merger saw a network of 41 branches consolidated into 24 branches. Ultimately, FINSAC’ stake in UBJ increased to 99.4 per cent. In 2001 the RBTT Financial Group acquired FINSAC’s shareholding in UBJ. RBTT owns banks in Guyana and Suriname, as well as others in the Caribbean, and originally was the Royal Bank of Trinidad and Tobago, a subsidiary of Royal Bank of Canada, but now is independent.

If one examines the market share of the three banks that were foreign-owned before the crisis, all grew faster than the intervened banks, and all gained at the expense of the formerly government-owned National Commercial Bank.

58

Table: Following the crisis Jamaica’s banking system is now almost entirely foreign-owned 1998 2001 Growth Bank Assets Per cent Assets Per cent Per cent Bank of Nova Scotia 57,087 34 87,279 36 53 Citibank 5,187 3 9,285 4 79 Canadian Imperial 11,138 7 16,618 7 49 Trafalgar Commercial 858 0.5 First Global Bank 2,991 1 249 Citizens Bank 12,870 8 Eagle Commercial Bank 3,295 2 Island Victoria Bank 2,504 1.5 Workers Savings & Loan 4,836 3 RBTT 33,298 14 42 National Commercial Bank 72,281 43 90,349 38 25 TOTAL 170,058 100 239,820 100 41 Per cent foreign 43.2% 98.8% Notes: Bold indicates assets in foreign-owned banks; Italics indicates banks with FINSAC intervention. Source: Bank of Jamaica

59

Appendix 7: Malaysia

From 1966 on, the Malaysian government barred existing foreign banks from opening new branches. It also limited any newly entering foreign banks to one branch. Further entry restrictions followed. The last foreign bank allowed to enter was Bank of Nova Scotia in 1973. In 1979 the government also issued its last new domestic banking licenses. In 1989 the government passed a law that required all foreign banks to convert their branches into domestically incorporated subsidiaries within five years. The government also limited foreign shareholdings in individual domestic banks to 10 per cent for an individual and 20 per cent for a corporation, and aggregate ownership to 30 per cent. The government has higher capital requirements for foreign banks, restricts them from opening branches (including ATMs—automated teller machines), and bars them from doing business with state-owned enterprises. The government did create an International Offshore

Banking Center in Labuan in 1990. Still, by definition, Offshore Banking Units have little access to the local market.

In 1998 the government took a number of measures to deal with the banking sector. It established Danamodal Nasional to recapitalize the banking system and Danaharta Nasional, the national asset management corporation, to remove non-performing loans from banks’ balance sheets. It also announced a plan to consolidate the 51 domestic banking institutions into just ten banking groups around 10 anchor banks by end-2001. The ten anchor or merged banks are

Maybank, Alliance Bank, Bumiputra-Commerce Bank, Affin Bank, Public Bank, Southern Bank,

RHB-Utama Bank, EON Bank, Arab-Malaysian Bank and Hong Leong Bank. Reportedly, the ruling United Malays National Organization held back consolidation while ensuring that trusted members of the party controlled all 10 anchor banks. Negotiations for the last merger, that of RHB and Utama Bank, only finished in 2002. Furthermore, some of the 10 anchors are not strong and further mergers are likely.

60

In March 2001, the government adopted a 10-year plan for banking sector reform.

Corporations’ and individuals’ ownership stakes in banks will not be allowed to exceed 20 per cent

and 10 per cent respectively. New foreign banks will be allowed to enter only after 2007. Malaysia

takes the view that it has no commitment under World Trade Organization accession to open its

domestic banking sector to more foreign competition and wishes to improve the efficiency and

effectiveness of the financial sector before opening it up to greater foreign competition. In 2000

there were 14 foreign banks with approximately 140 branches between them. (The Bank of China

was allowed to return in 2000 after a 41 year absence, but this represents an exception.) In 2002 two

Singaporean banks merged, resulting in the merger of their subsidiaries in Malaysia. The merger of

OUB Malaysia into UOB Malaysia made UOB Malaysia the largest foreign bank in Malaysia.

Currently foreign banks account for 24 per cent of banking system assets. Apparently the foreign banks gained several percentage points during the crisis and immediately thereafter but perhaps the domestic banks are now regaining lost ground.

Table: In Malaysia the crisis hardly increased the limited foreign bank presence (RM Millions & % yony) Domestic Foreign Total Frn/Total %∆ %∆ %∆ 1996 279,986 80,141 360,127 22% 1997 376,434 34.4% 103,815 29.5% 480,249 33.4% 22% 1998 352,765 -6.3% 101,748 -2.0% 454,513 -5.4% 22% 1999 373,646 5.9% 109,961 8.1% 483,607 6.4% 23% 2000 388,727 4.0% 123,98812.8% 512,7156.0% 24% 2001 398,156 2.4% 131,580 6.1% 529,736 3.3% 25% 2002 425,792 6.9% 137,158 4.2% 562,950 6.3% 24% Source: Bank Negara Malaysia

61

Appendix 8: Mexico

From the late 1920s to the mid-1990s the only foreign bank in Mexico has been Citibank, whose tightly circumscribed operations were grandfathered when Mexico closed. In 1982 the

Mexican government nationalized the banking system and subsequently consolidated the 68 domestic banks into 18. In 1990 the government started to liberalize and then in 1991 and 1992 it privatized the banks though it excluded the possibility of foreign control (Minushkin 2002).

Banking law limited foreign ownership to a maximum 20 per cent position in any one bank. A period of “hypercompetition” apparently followed as the newly privatized banks sacrificed profitability to gain market share and position (Gruben and McComb 2003).

This hypercompetition left the banks vulnerable when the 1994 Peso crisis devastated the economy. That, and insider and crony lending resulted in many banks ending up back in the government’s hands (Calomiris et al., 2003, Honohan 2003, and La Porta et al., 2003). The subsequent 1995 banking reform reduced the restrictions on foreign ownership and authorized the

Ministry of Finance and Public Credit to allow foreign financial institutions to acquire a controlling interest in those banks whose capital represented less than 6 per cent of the total net capital of the

Mexican banking industry. In 1998 the Mexican Congress finally approved foreign ownership of up to 100 per cent in Mexican banks.

In 1993 Scotiabank (Bank of Nova Scotia) acquired 8.5 per cent of Grupo Financiero

Inverlat, Mexico’s seventh largest bank. In 1996 Scotiabank received a management contract from the government. Then in 2000, Scotiabank increased its ownership to 55 per cent and took control of Inverlat. It is currently (2003) negotiating to buy the government’s remaining 36 per cent stake.

BBVA (Banco Bilbao Vizcaya Argentaria) started its Mexican acquisitions in 1995 by buying

Banco Mercantil Probursa. In 1996 it acquired Banca Cremi and Banco Oriente. Then in 2000

BBVA Probursa merged with Bancomer (Banco Comercial Mexicano), Mexico’s second largest

62 bank, to form Grupo Financiero BBVA Bancomer. Bank of Montreal agreed in 2001 to sell its shares in Grupo Financiero BBVA Bancomer to BBVA. (Bank of Montreal acquired an early 20 per cent position in Bancomer but then chose not to increase its stake.)

In 1997 Citibank acquired Banca Confia and bid unsuccessfully for Banco Serfin. Then in

2001 Citibank acquired Banamex (Banco Nacional de Mexico). Banamex will consolidate both

Banca Confia and Citibank of Mexico. Before BBVA acquired Bancomer Banamex was Mexico’s largest bank. The merger with Citibank’s other operations returns Banamex to the top position.

BSCH began with the acquisition in 1997 of InverMexico, owner of Banco Mexicano, then

Mexico’s fourth largest bank. In 2000 it acquired Banco Serfin, Mexico's oldest (founded in 1864) and third largest bank. In doing so it beat out a bid from HSBC. Late in 2002, Santander sold 24.9 per cent of Santander-Serfin to Bank of America. Both banks wish to use the alliance to tap remittances by Mexicans in the US.

In 2002 HSBC bought all the shares of Bital (Banco Internacional), including Santander’s 27 per cent and ING Group’s 19 per cent. HSBC added another US$800 million into Bital to increase its capital.

Currently, the top three banks in Mexico (Banamex, BBVA Bancomer and Santander Serfin) and the fifth (Bital) are all foreign-owned and account for over 80 per cent of banking system assets.

The degree of foreign ownership is as high as in some Eastern European countries and the ownership is concentrated in far fewer banks.

63

Appendix 9: Norway

In 1973, after the Ministry of Finance ruled that representative offices did not come under

the banking law and hence foreign banks did not require permission to open one, several foreign

banks entered. Still, in 1974 the Government formally rejected applications from foreign banks to

establish branches or subsidiaries. In 1984 the Norwegian government substantially deregulated the

domestic market, allowing foreign banks to establish subsidiaries in Norway and Norwegian banks

to establish branches abroad. Still, Norway forbade foreign banks to operate as branches of the

parent and also limited foreign banks to at most small shareholdings in existing domestic banks. In

the period of competitive turbulence that followed deregulation, several existing domestic banks

merged and investors created several new domestic banks. Between 1985 and 1996, eleven foreign

banks entered, of which five still remained in 1996 (Tschoegl 2002a).

Deregulation led to an overall doubling of loans outstanding (without a commensurate

increase in staff) as banks pursued market share (Lindblom 1994). However, by 1987 it was already

becoming apparent that many of these loans were of low quality. The banks' loan loss provisions

began to rise markedly as the government imposed stricter accounting standards.

By 1990, it was apparent that a crisis had arrived. In response, the government agreed to permit foreign banks to acquire smaller, failing domestic banks. Banque Nationale de Paris took over one and Sweden’s Handelsbanken took over two and parts of a third.

The largest and third largest Norwegian banks, Den norske Creditbank (DnC) and Bergen

Bank (BB), merged under the name Den norske Bank (DnB). By the end of 1991, the government had come to wholly-own Christiania Bank og Kreditkasse (CBK) and Fokus Bank (by then the second and third largest banks) and a majority share in DnB. The government liquidated one smaller bank, Norion Bank and demanded the restructuring of the banks it owned. During this

64 period, the market share (in terms of assets) of the foreign banks fluctuated between three and four per cent, moving contra-cyclically (Engwall et al. 2001)

After recapitalization, the Norwegian banks recovered quickly. In 1993 Norway acceded to the European Economic Area and agreed to liberalize further its regulations on entry. In 1994,

Norway finally authorized foreign banks to enter via branches of the parent. Three foreign banks converted their subsidiaries to branches and one entered with a branch. In 1999, Den Danske Bank acquired Fokus Bank, in the process both outbidding Handelsbanken and disrupting the government’s plan to merge Fokus with CBK and Postbanken. Ultimately, the Norwegian government merged Postbanken into DnB, with the result, inter alia, that the government’s ownership of DnB rose to 61 per cent.

Handelsbanken did make a successful bid for Bergensbanken, which increased

Handelsbanken's assets in Norway by 50 per cent. Handelsbanken pledged to keep Bergensbanken separate, to maintain existing management, and to transfer Handelsbanken's ship finance and Bergen branch to Bergensbanken. Handelsbanken then made a bid for CBK, or at least the portion not in the hands of the Norwegian government, but withdrew its bid in the face of government opposition.

Eventually (October 2000), Nordea, the Finnish-Swedish bank, made a successful bid for CBK.

Nordea then made a compulsory purchase of the remaining shares, delisted CBK from the Oslo

Stock Exchange, and acquired a secondary listing for itself.

In 2001 the Norwegian government sold 13 per cent of DnB, reducing its stake to 47 per cent. Then in 2003 DnB acquired Nordlandsbanken and merged with Gjensidige NOR Sparebank, the banking arm of the Gjensidige NOR financial services group. The Norwegian government will probably have to increase its investment if it is to retain a 34 per cent stake in DnB after the merger.

65

Appendix 10: Pacific Islands

In Fiji, Nauru, Papua New Guinea, the Solomon Islands and Vanuatu major government-

owned banks failed in the 1990s. The reasons had more to do with political lending than any natural

catastrophes or demand shocks. Since then, most the local governments have privatized, generally

to foreigners, most of the banks that they had acquired or established. Australian, French and

Guamanian or Hawaiian banks dominate banking throughout the Australian, French and American

spheres of influence, respectively. In general, branches and subsidiaries of foreign banks account for

the majority of assets in the region’s banking systems.

Following independence, governments tended to wish to localise banking. This took the

form of creating locally incorporated banks often on the foundations of a foreign bank, with the

government taking a minority or majority position in the new institution. Generally, the

governments began by giving foreign banks (in some cases the former owner) a management

contract. As these contracts expired, and as governments recognised the political utility of directing

lending, the governments took over the management themselves.

The Fijian government established the National Bank of Fiji (NBF) in 1974 on a foundation

of the Savings Bank of Fiji. Until the 1980s, Commonwealth Bank provided technical assistance and

seconded some personnel. Then in 1987 the government replaced the CEO with someone who had

little experience and knowledge of running a bank, and who announced that the National Bank of

Fiji wanted to be a “big player” amongst Fiji’s banks (Siwatibau 1996). By mid-1995, NBF was

running bad and doubtful debts that eventually amounted to US$220 million, or 8 per cent of Fiji's

GDP. The debtors were mostly prominent citizens who were never prosecuted; they also lost no

collateral, the bank having never asked for any. The government responded to public outrage and

media scrutiny by minimizing the seriousness of the situation, resisting demands for a public

enquiry, and by attempting to muzzle the press. Finally, in 1996-7 the government split NBF into

66

two parts, the Asset Management Bank, which would be responsible for winding down the

commercial and non-performing loans, and NBF proper. The restructuring also involved a more

than 50 per cent reduction in the number of branches, particularly in rural and outlying areas. In late

1998 the government drew on the national provident fund to make good the financial shortfall.

Then in 1999, it privatized NBF by selling 51 per cent to Colonial Ltd., an Australian company that

had been operating in Fiji as a life insurance company since 1876. NBF’s trading name became

Colonial National Bank (CNB). In 2000, Commonwealth Bank of Australia bought Colonial and so

CNB. Plans to sell the government’s remaining 49 per cent of CNB are on hold.

In 1968 the government of Nauru, which had achieved independence one year earlier,

authorized Bank of New South Wales (BNSW) to open a branch that would function also as the

official bank in the Republic. The agreement gave BNSW a 10-year monopoly but included the

proviso that the government could take over the branch at any time after 15 months had elapsed. In

1976 the government exercised its option by establishing the Bank of Nauru to take over the branch.

The original agreement committed BNSW to provide banking management and service assistance

but instead the government turned to Barclays Bank, which also took a 10 per cent stake in the

bank. In 1985 Barclays Bank sold its stake to the government. The Bank of Nauru collapsed in

1995 but may be limping along though it is not clear if Nauru has a functioning government at this

time.

In 1974, as independence for Papua New Guinea (PNG) approached, Commonwealth Bank

Corporation (CBC) handed over the bulk of its PNG operations to the newly created and

government-owned Papua New Guinea Banking Corporation (PNGBC). When the IMF and the

IBRD made privatization of PNGBC and several other government-owned corporations a condition for further assistance, PNG-owned Bank of South Pacific (BSP) acquired 51 per cent of PNGBC in

67

2002. Later BSP merged PNGBC, giving the government a 25 per cent share in BSP in return for

its 49 per cent stake in PNGBC.

National Bank of the Solomon Islands (NBSI) has a 42 per cent share of the market, the rest

being in the hands of two Australian banks. In 1981 CBC, which had entered in 1951 by

establishing a branch, transferred its operations to the newly-incorporated NBSI, in which it took a

51 per cent share, the government holding the rest. Bank of Hawaii bought CBC’s shares in 1994

but in 2002 sold them to the government. NBSI may be close to insolvency as political instability

has left the economy in tatters.

The government established National Bank of Vanuatu (NBV) in 1991. In 1998 it

restructured NBV after it had sustained losses, transferring to it assets of the Development Bank of

Vanuatu. The government established an Asset Management Unit to recover and manage more than US$25 million in debts for the Vanuatu National Provident Fund, the Development Bank of

Vanuatu, and NBV. The government put the Asset Management Unit under new management in

2001. In 2001, NBV announced further major losses. Foreign banks account for in excess of 90 per cent of banking system assets.

68

Appendix 11: Tanzania

In 1967 the Government of Tanzania nationalized all banks in the country. The government

consolidated the assets and liabilities of all seven foreign and two local banks into the government-

owned National Bank of Commerce (NBC; est. 1965). This bank ended up lending to series of

insolvent parastatals, backed by a government that had negligible financial resources (Harvey 1991).

For instance, in 1987, 30 per cent of its loans were to the National Milling Corporation, which was

suffering large losses (Harvey 1991).

In 1991, the government liberalized the banking and financial sector, permitting the entry of

private banks. The government restructured and recapitalized the state-owned banks, NBC and Co-

operative and Rural Development Bank (CRDB). The first accounted for 90 per cent of all deposits

and the latter 5 per cent. The government admitted that lack of competition and government

interference had had an adverse impact on the sector and that it had failed to operate the banks.

Standard Chartered Bank Tanzania, a subsidiary of Standard Chartered Bank of the UK

Limited, was the first foreign bank to commence business after liberalization. It started operating in

Tanzania in 1993, though the parent bank had operated in Tanganyika from 1917 to 1967. As of

mid-2000, Standard Chartered was the fifth-largest bank in the country (in terms of total assets) and

had six branches.

Citibank commenced its business operations in Tanzania in 1995. The bank is engaged in

wholesale commercial banking business and caters to the institutional clients. As of mid-2000 it was

the third-largest bank in the country. The sixth largest bank is Stanbic Bank Tanzania, which is the

subsidiary of Standard Bank Group of South Africa and which also entered in 1965. Stanbic entered

by acquiring the Tanzanian subsidiary of Meridian Bank BIAO which had recently failed.

In 1996 the government restructured and recapitalised CRDB to form CRDB (1996) Ltd, which it then privatized. The Danish International Development Agency is the largest single

69 shareholder and has also provided technical, managerial and financial support. As of mid-2000

CRDB (1966) was the second largest bank in the country, ranking just behind NBC.

In 1997 the government reorganized NBC and early in 2000 it sold 55 per cent of the shares to Amalgamated Banks of South Africa (ABSA) and 15 per cent to the International Finance

Corporation, while retaining the remainder. NBC is still the largest bank and has 34 branches. Prior to the sale, the government recapitalized the bank and spun off a microfinance unit: National

Microfinance Bank Limited (NMB). NMB caters for SMEs and has over 90 rural branches. At mid-

2000 it was the fourth-largest bank.

As of mid-2000 foreign-owned banks accounted for over 55 per cent of the assets in the

Tanzanian banking system. The top 6 institutions, of which four are foreign-owned, accounted for

89 per cent of the assets in all banks and non-bank financial institutions in 2000.

70

Appendix 12: Thailand

From the 1960s, the Bank of Thailand de facto barred the entry of any further foreign banks with an occasional exception. The government limited those few foreign banks that had managed to open before the prohibition to operating only one branch office. In 1993, the Bank of Thailand granted 47 licenses for operations under the newly created International Banking Facility. These licenses were the first granted to foreign banks since 1978. Over half of the licenses went to local banks and to branches of foreign banks in Thailand. However, recipients also included 20 foreign banks with no branch operation in the kingdom.

The domestic banks in Thailand were not well run. They engaged in liability and currency mismatching, borrowing long and lending short, and borrowing in US$ and lending in Baht.

Furthermore, the banks lent on the basis of cronyism, rather than credit assessment

(Wiwattanakantang 2002). What facilitated the cronyism was that 12 of the 15 commercial banks were family owned.

Once the crisis began, rather than shutting the banks down, the Bank of Thailand gave them two years to increase their capital. Still, dealing with the crisis involved the government taking over six banks. It created Radanasin Bank in 1988 as a “good bank” to take over the assets of the 56 finance companies that it closed. The government now owns three banks, which account for about

27 per cent of banking system assets. Foreign banks acquired three banks from the government.

United Overseas Bank acquired Radanasin Bank, which had itself, under duress from the regulatory authorities, taken over the defunct Laem Thong Bank. Standard Chartered acquired Nakornthon

Bank. In addition, Development Bank of Singapore acquired Thai Danu Bank and ABN-AMRO purchased the Bank of Asia, on a buy-now, set-the-price-later plan. As a result, the share of foreign owners in banking system assets approximately doubled but, relative to the situation in Latin

America, remained small.

71

Even in several banks where the owning families managed to retain control, foreign ownership increased when the banks sought additional capital. At Bangkok Bank, Bank of Ayudhaya, Siam

Commercial Bank, Thai Farmer’s Bank foreign ownership was 25 per cent in 1997; it was 48, 40, 49 and 49 per cent respectively at the end of 1999 (Hewison 2001).

Furthermore, some additional sales fell through. In late 2000 HSBC withdrew from negotiations to buy a 75 per cent stake in Bangkok Metropolitan Bank (BMB) and efforts to sell

Siam City Bank (SCIB) to Newbridge Capital also collapsed. The government merged BMB into

SCIB in 2002 and may yet privatize SCIB eventually. In 2003 negotiations between the government and ANZ Bank over the privatization of Thai Military Bank failed.

Table: Despite a doubling in its share, in Thailand the foreign bank presence remains limited 1996 2001 Branches of foreign banks 14 18 Local banks w/majority foreign ownership 0 4 Number of local banks 15 12 Assets in branches or subsidiaries of foreign banks 8.5% 17.5% Source: Thai Ministry of Finance.